learning - Utrecht University Repository - Universiteit Utrecht

3 downloads 0 Views 3MB Size Report
teachers' regulation of learning in dual learning programmes. 77. Chapter 5 ..... Most models of self-regulation also share the same four phases, sometimes also ...
significantevaluated increase explain means pected kinds fragments processes indication conceptual questions predicted female views

orientations role

programme

detail consisted

erpretativeassociatesmultilevel feedback explored

tern-track starting

became

rubrics

interventions

problems classroombe

student teachers’ preferences occasions

quality appendix

carried

profession

previous sources inter-rater

context

low-scoring

attempt

factors

focused

instructional

tracks

aptitude

ability ideas analysis adaptive

nonlinear

rganised

learning provided

concerns

oriented

conclude applied informal

become

tional

pert

mes

ion

positively

social

nt

individual reactive

overview

identity

whereby

pre-service

p

inv

collection

positioned

frame

publishers

d

regarding par seconda

discussion

interns argument

generally

desc

enactm

profess

researchversion va

questionnaire influence

model presented

outcomes iltp

opportunity

job-track

decrease

selection

sage

project

learning rep monitoring liked

argumentation

years

emerging

forethought

handbook aggregated

identify

namely

publications

reflect component

definition one-third

small-scale

vering

traditional

patterns

completed

variables institutes

do

sugg framework correlations assumed importa

resulted

improving

interpretation structure discriminate reasons

eptions

disso

term chapte

knowledge significantly experiences respectively

rema

concurrent larger conditions

variances

implications

self-regulated amount conclusion

ctor regular

sco already

situations

personal relation extent various choosing aim academic self-directed missing

hoice

classified

reliable

goal-setting achievement

ales

sam

categories cons concept goal international variation assessment

cross

ve

decades

teach

included

increased

similarities confirmed

indicates

creasing

dim

detailed instruction variety self-effica

growth

varied

collected

directed

criteria

acknowledge

styles

regulatio

literature explanation variance

discus

slope following datasuggested Maaike Endedijk con workplace e predictive tualised meaning-oriented ethods

tistics

situated

quadratic

psychometric exploring behavior factor

strategies competence attention

getting

frequencies

answering kappa conclusions referr

source coherent shared

interviews

Student teachers’ self-regulated learning

Maaike Endedijk

IVLOS Series The IVLOS-series is published by IVLOS Institute of Education of Utrecht University (Instituut voor Lerarenopleiding, Onderwijsontwikkeling en Studievaardigheden). The purpose of this series is the dissemination of results of research to enhance the quality of education. The members of the editorial board are: Prof. dr. A. Pilot Prof. dr. P. R. J. Simons Prof. dr. J. D. H. M. Vermunt Recent publications in this serie are: I. Zitter. Designing for learning. Studying learning environments in higher professional education from a design perspective. M. N. Rosenfeld. Developing teacher sensitivity to individual learning differences. Studies on increasing teacher effectiveness. M. J. J. Coenders. Leerarchitectuur.Een exploratief onderzoek naar de relatie tussen ruimte en leren in werksituaties en het ontwerpen voor leren dichtbij de praktijk M. Moonen. Testing the multi-feature hypothesis. Tasks, mental actions and second language acquisition. Ä. Leijen. The reflective dancer. ICT support for practical training. M. P. Nguyen. Culture and cooperation: Cooperative learning in Asian Confucian heritage cultures – The case of Viet Nam. A. Hoekstra. Experienced teachers’ informal learning in the workplace. J. van der Pol. Facilitating online learning conversations: Exploring tool affordances in higher education. S. F. Akkerman. Strangers in dialogue: Academic collaboration across organizational boundaries. This research was carried out in the context of the Interuniversity Center for Educational Research (ICO) and funded by the Netherlands Organisation for Scientific Research (NWO) (Project no. 411-03-304). Endedijk, M. D. Student teachers’ self-regulated learning (Zelfregulerend leren van docenten in opleiding) Proefschrift Universiteit Utrecht – Met literatuuropgave – Met samenvatting in het Nederlands Cover design, graphics & lay-out: Wietze Nijdam ISBN 978-90-393-5339-4 © 2010, Maaike Endedijk All rights reserved. No part of this thesis may be reproduced, stored in databases or retrieval systems, or transmitted in any form or by any means, electronical, mechanical, photocopying, recording or otherwise, without the prior written permission of the author.

Student teachers’ self-regulated learning

Zelfregulerend leren van docenten in opleiding (met een samenvatting in het Nederlands)

Proefschrift

ter verkrijging van de graad van doctor aan de Universiteit Utrecht op gezag van de rector magnificus, prof.dr. J.C. Stoof, ingevolge het besluit van het college voor promoties in het openbaar te verdedigen op vrijdag 2 juli 2010 des middags te 2.30 uur

door

Maaike Dorine Endedijk

geboren op 30 april 1982 te Westervoort

Promotoren Prof. dr. J.D. Vermunt Prof. dr. M. Brekelmans Prof. dr. N. Verloop Co-promotor Dr. P.C. Meijer

Contents Chapter 1 General Introduction

7

Chapter 2 The nature of student teachers’ regulation of learning in a dual learning programme

27

Chapter 3 Student teachers’ self-regulated learning: An examination of regulation patterns and conceptions of learning in a dual learning programme

49

Chapter 4 The development and evaluation of an instrument to measure student teachers’ regulation of learning in dual learning programmes

77

Chapter 5 Student teachers’ development in self-regulated learning throughout a dual learning programme

109

Chapter 6 General conclusions and discussion

143

References

173

Samenvatting (Dutch summary)

189

Publications

203

Curriculum Vitae

207

Dankwoord

209

outcomes

stages

paths

starting

observation regarded

teaching

education concerns kinds

responsibility

environment

decades thesis

detailed

positive

various

described

relatively

extent

learners

tillema

independent complex

goals

following

netherlands

cognitive

tracks

conceptions

reproduction-oriented

srl

combination

institutes

small-scale

becoming

exists

challenge

carried

knowledge conceptual interventions situations job-track hammerness orientations meaning-oriented enactment professional

self-regulated classroom

directed

university

specificanswering focus

schools

conceptualised

means

hagger

processes

oosterheert styles

one-year

promote

nowadays

reflect

patterns

activities mcintyre

profession

focused

individual

framework already

understanding prospective

provided

expert

monitoring expected

student teachers

identified

views lacking

lifelong

multiple

shared

integrated vermetten

competences preferences

marton

dimensions resulted

traditional

lonka

increase

dissertation

school-based

makes

adaptive

involves

simultaneously

experiences

colleagues

association

conclude

problems

interview internship

learning review

perspective

quality

dynamic structured reference

relation

stable

dual

teacher

models

secondary

shortage

research

information

constructs

hollingsworth

inventories

kremer-hayon

different

levels

important

vermunt

theories

assumption qualitative academic dependent

applying

upper

approach feedback

mutton

phases

including

programme

pre-service confirmed skills

students

development verloop

definition

versus

clarke

organised

overview

designed experienced personal

metacognitive

across

description

namely

regulation

task

terms

chapter

conclusions

data

years

van petegem

perry

insights

assumed

attempt

closed

strategies

context

quantitative

frame

defined

aspects motivation

1 General introduction “Becoming a teacher is not only copying from what you read in the books or from what you see other teachers doing. Becoming a teacher is continuously experimenting with what fits you best as a teacher.”

“You have to connect systematically all your knowledge with the feedback you are getting, to get new insights. Yes, I really think, teaching is also being a learner. You really have to keep learning till the final, final end.” (Citations from two student teachers who participated in the studies described in this dissertation)

The focus of this thesis is on student teachers’ self-regulated learning in a dual learning programme, in which learning in practice and at university are combined. In learning to teach, three questions are important: What do teachers need to learn, how do they learn and develop, and how can this learning be stimulated? Much of the research in the field of teacher education is concerned with the relation between the first and last question: How actions, knowledge, and beliefs of student teachers or experienced teachers change as a function of different teacher education programmes (Cochran-Smith & Zeichner, 2005). While a lot of attention has been given to how their pupils learn in schools, the second question about how teachers learn is much less addressed in research (Beijaard, Korthagen, & Verloop, 2007). Pre-service programmes, especially one-year post-graduate teacher education programmes only form a very time-limited intervention in the whole professional development of a teacher. Therefore, it is extremely important that these programmes support the development of student teachers’ conceptions and skills necessary for their further professional development as 7

a teacher. The quality of student teachers’ learning depends for a significant part on the ability to control and steer their learning (Niemi, 2002). Our assumption is that deeper understanding of the concept of self-regulated learning in learning to teach is necessary in order to design powerful teacher education programmes for supporting this development. Currently, detailed descriptions of how and to what extent student teachers regulate their learning to teach, as well as how it develops throughout a dual learning programme is lacking. Therefore, this thesis will attempt to broaden the understanding of the nature and development of student teachers’ self-regulated learning in a dual learning programme. In this General introduction, first the relevance of self-regulated learning for the three questions about learning to teach as posed above, will be addressed. Furthermore, an overview of the concept of self-regulated learning will be given, including how it has developed historically and how it has been studied so far in the context of learning to teach. This overview will lead to the specific research questions for this thesis.

1.1 Research on learning to teach 1.1.1 What teachers need to learn Feiman-Nemser (2008) has conceptualised four general aspects that have to be learned as a teacher: learning to think, know, feel, and act like a teacher. Thinking like a teacher refers to critically examining one’s beliefs, interpreting teaching activities in a pedagogical framework and reflecting on and adjusting one’s practice. Knowing like a teacher comprises the different kinds of knowledge for teaching and knowledge of teaching, which include, for example, subject matter knowledge and pedagogical content knowledge (Shulman, 1986; Van Driel, Verloop, & De Vos, 1998). Feeling like a teacher includes forming a professional identity (Beijaard, Meijer, & Verloop, 2004; Rodgers & Scott, 2008) and to act like a teacher relates to the development of a repertoire of skills and strategies and to know when to use them. The biggest challenge after all is to integrate these four aspects of learning to teach (Feiman-Nemser, 2008). And for that, in our opinion, teachers also have to learn to learn as a teacher. Even for very experienced teachers, teaching will never be complete routine, because, for example, students are never the same, subject matter knowledge develops and demands from society change. Therefore, to become 8

Chapter 1

a professional, teachers need to develop what Hammerness and colleagues call ‘adaptive expertise’ (Hammerness et al., 2005). Adaptive expertise consists of two dimensions: efficiency and innovation. Teachers who are adaptive experts are able to balance these two dimensions in order to do a variety of teaching activities at the same time without too much effort and also to rethink their routines and develop new practices. Besides applying routines, they are able to improve over time and appreciate the value of actively seeking feedback from many sources in order to solve problems that arise while continuing to meet the needs of their students (Hammerness et al., 2005). The importance of active lifelong learning as part of being an expert teacher has often been mentioned (Beijaard et al., 2007; Bolhuis, 2003; Clarke & Hollingsworth, 2002; Hammerness et al., 2005). This lifelong learning, however, is not always a matter of free choice: external factors, such as educational reforms, new technologies, and new learning theories ask of the teacher to reconsider their ideas and change their practices (Beijaard et al., 2007; Butler, Lauscher, Jarvis-Selinger, & Beckingham, 2004). Also in-service professional development programmes designed to support teachers with these changes take more and more into account the responsibility of the teacher for shaping their own learning (Butler et al., 2004; Clarke & Hollingsworth, 2002; Niemi, 2002). According to Randi (2004), this makes it difficult to imagine an effective teacher who has not developed self-regulated learning strategies. In conclusion, adaptive expert teachers have not only learned how to think, know, feel, and act like a teacher, but also know how to remain an agent of their own learning process in times of rest and change.

1.1.2 Teacher learning and development During the last decade, an important shift occurred in research into learning to teach from researchers focusing on what teachers need to know, care about, and be able to do, to questions as to how teachers acquire, generate and learn to use knowledge in teaching (Feiman-Nemser, 2008). The current literature describes explicitly that learning to teach is a complex, multidimensional, idiosyncratic and context-specific process (e.g., Clarke & Hollingsworth, 2002; Mutton, Burn, & Hagger, 2010). This might explain why research on learning to teach is generally qualitative in methodology, based on small groups of teachers in specific contexts. The apparent difficulty in generalising about these studies resulted in limited knowledge of general processes of teacher growth (Cochran-Smith & Zeichner, 2005; Kagan, 1992). A general description 9

of teacher learning, such as given in the Handbook of Research on Teacher Education (Cochran-Smith, Feiman-Nemser, McIntyre, & Demers, 2008) is not different from the general social-constructivist idea of learning, namely that learning is not a passive process of absorbing information, but influenced by what we already know and by the social and cultural contexts (FeimanNemser, 2008). Paradoxically, on the one hand learning to teach has been described in terms of general learning theories, which makes teacher learning not so different from student learning (Hammerness et al., 2005). On the other hand, the individual backgrounds and dynamic contexts in which teachers learn vary a lot, which makes it hard to arrive at general principles and processes which describe specific teachers’ learning processes adequately. Despite the context-specific nature of learning to teach, some general models have been developed to describe different stages in teacher development. Well-known examples are the model of Fuller (Fuller & Parsons, 1972), describing the shifting concerns of new teachers from a focus on their own survival in the classroom to a focus on their students’ learning needs, and the five-stage theory of the development of expertise of Berliner (1988). This last theory showed how teaching changes from rationally applying rules to a fluid and effortless way of teaching in which choices are made more intuitively. Some teachers never reach the final stage, and if they do, they need at least five to seven years of experience (Berliner, 2001). These stage theories have helped to describe possible trajectories in teacher development, but they do not explain how teachers develop towards more advanced stages and research has shown that these paths in learning to teach are often more complex than linear changes through fixed stages (Hammerness et al., 2005). In their review on how teachers learn and develop Hammerness and colleagues (2005) have identified three challenges for the development towards an adaptive expert: the problem of what Lortie (1975) has called the ‘apprenticeship of observation’, the problem of enactment, and the problem of complexity. The apprenticeship of observation refers to the preconceptions of prospective teachers about how teaching works, based on their long experience as a student in the classroom. These preconceptions about teaching and learning can be hard to identify and change and may hamper their development as a teacher according to the contemporary views of learning. The problem of enactment describes the difficulty to put intentions into action (Hammerness et al., 2005). Enactment and reflection are defined by Clarke and Hollingsworth (2002) as the core mediating processes which teachers need in 10

Chapter 1

order to connect their different domains of learning. This enactment is difficult for teachers since teaching is a highly complex task which involves doing things simultaneously by making use of different kinds of knowledge to serve a large group of students, all with different needs (Hammerness et al., 2005). The last challenge describes the complexity of teaching: teaching involves reaching multiple goals in relation to a diverse group of students, which requires multiple kinds of knowledge to be used in an integrated way. According to Hammerness et al., teachers need a metacognitive approach to deal with this complexity and to take control of their own learning in the complexity of the classroom. In summary, no matter how good student teachers’ preparation is and how well they have done in their initial practice, the knowledge, skills, and attitudes needed for being an expert teacher cannot be fully developed in preservice programmes (Feiman-Nemser, 2001; Hammerness et al., 2005). Furthermore, the specific challenges of teacher learning in a complex and dynamic learning environment of everyday practice shows that a metacognitive approach is an essential skill for teacher learning (Hammerness et al., 2005). According to the study of Niemi (2002), the quality of learning is dependent on the ability of the student teacher to steer his or her own learning and to learn to reflect on and control his or her own learning processes. In conclusion, to prepare student teachers for their further development as a teacher, one of the main purposes of teacher education is that student teachers become able to focus on themselves as learners and be equipped for selfregulated learning (Conway & Clark, 2003; Hagger, Burn, Mutton, & Brindley, 2008; Hammerness et al., 2005; Kremer-Hayon & Tillema, 1999).

1.1.3 Learning to teach in different pre-service programmes Many traditional pre-service teacher education programmes have been criticised over the past decades for separating theory from practice, offering little opportunity for self-regulated learning, and having no shared or too conservative conceptions of teaching and learning (Berliner, 2000; DarlingHammond, 1999; Darling-Hammond, Hammerness, Grossman, Rust, & Shulman, 2005; Feiman-Nemser, 2001; Hagger et al., 2008; Kremer-Hayon & Tillema, 1999). Student teachers often felt after these pre-service programmes unprepared to start teaching. They faced a severe practice shock and had problems surviving in the classroom, because they did not know exactly how to apply in the classroom what they had learned at the teacher education 11

institute (Darling-Hammond, 1999; Hagger & McIntyre, 2000; Korthagen, in press). Apparently, these traditional programmes were not designed well enough to promote the complex process of learning to teach. These critiques led to changes towards more school-based teacher education programmes, starting in the UK in the beginning of the 1990s (Hagger et al., 2008; KremerHayon & Tillema, 1999; Maandag, Deinum, Hofman, & Buitink, 2007; McIntyre & Hagger, 1992). This resulted in an increase of time spent by prospective teachers in the classroom and a more tight connection between field experience in schools and method courses at higher education institutes (Cochran-Smith & Zeichner, 2005; Edwards & Mutton, 2007; Hagger et al., 2008) In contrast to the UK, the move towards school-based teacher education in The Netherlands has developed more quietly and more recently (Maandag et al., 2007; Ten Dam & Blom, 2006). One reason is that The Netherlands have a longer tradition of organising the content of the curricula around the tasks of teachers and giving student teachers time to spend a substantial part of their teacher education in schools (Ten Dam & Blom, 2006). However, the still increasing shortage of teachers was a reason for the government to stimulate more cooperation between schools and teacher education institutes (Maandag et al., 2007). The one-year post-graduate teacher education programme for upper secondary education in The Netherlands can be best described as a dual learning programme, in which learning at university is combined with learning in practice. These programmes include models in which the student teacher is employed as a teacher and the student teachers have all the responsibilities of a regular teacher (Buitink, 2009), as well as models in which student teachers do an internship and are more gradually exposed to the teaching profession. Also, first experiments have been started in which the schools are getting more responsibility for the teacher education programme (Maandag et al., 2007). Although learning to teach in these school-based and dual learning programmes is more integrated with student teachers’ teaching practice than in the traditional programmes, an increase of practical experiences does not automatically lead to successful learning from these experiences (DarlingHammond, 1999). Student teachers also need the capacity to learn from these experiences and integrate them with theory. Two empirical studies have identified important individual differences in learning to teach in these dual learning programmes. The study of Oosterheert and Vermunt (2001) identified the following dimensions to be important for describing differences in the way 12

Chapter 1

student teachers learn to teach: the dimension meaning-oriented versus performance-oriented learning shows differences between student teachers in whether they learn by questioning and changing their frame of reference, or whether they are directed at improving performance through gathering practical suggestions. The dimension open versus closed learning shows differences in how student teachers approach their problems: whether they acknowledge their problems and try to find solutions independently or whether their problems remain more implicit and they need others to solve those (Oosterheert & Vermunt, 2001). Oosterheert (2001) suggested that a combination of an independent and meaning-oriented way of learning is most beneficial for preparing student teachers for lifelong learning. The study of Hagger et al. (2008), described the nature and extent to which student teachers learn from experience in teaching practice in five dimensions. These dimensions include: the intentionality of learning; the frame of reference that is used; the response to feedback; the attitude to context; and the aspiration for their own and pupils’ learning. Their small-scale study with four student teachers showed that although the student teachers claimed to learn from practice, they differed in the degree to which they learned in a deliberate way, extended their frame of reference, were able to make use of feedback, took own responsibility instead of blaming the context and had high aspiration for their own learning. Both studies of Hagger et al. (2008) and Oosterheert and Vermunt (2001) have shown that student teachers differ in the extent to which they learn in an active and independent meaning-oriented way. Hagger et al. (2008) stated that for student teachers who lack this capacity, the process of learning from experience can be seen as miseducative, since it reinforces the idea that one can learn to teach by a simple accumulation of practice. A more reactive way of learning makes it less likely that student teachers learn successfully from these contexts (Mutton et al., 2010). According to Mutton et al. (2010), the only way to prevent this is to help and teach prospective teachers how to take control of their professional development and how to continue learning, no matter how successful teachers might be in demonstrating their practical competence. Furthermore, dual learning programmes take into account the personal agendas of student teachers which they bring to the programme and in this way give a very active role to student teachers in designing their own learning path (Hagger & McIntyre, 2000; McIntyre & Hagger, 1992). Therefore, already during a dual teacher education programme student teachers need to be able to identify their learning needs 13

and to plan, monitor and evaluate their progress on their individual learning paths. To conclude, self-regulated learning is not only a desirable outcome of teacher education, but in dual learning programmes these conceptions and skills are necessary in order to learn from a combination of theory and practice.

1.2 Research on self-regulated learning As has been described above, self-regulated learning (SRL) is a very important aspect of becoming and being an adaptive expert as a teacher and a necessary skill for learning to teach in a dual learning programme. This means that selfregulated learning is a means as well as an end in pre-service teacher education. Below, an overview will be given of the concept of SRL and how it is studied so far in the context of learning to teach.

1.2.1 The scope of the concept of self-regulated learning Recently, Dinsmore, Alexander, and Loughlin (2008) have published a review of how metacognition, self-regulation, and self-regulated learning have been defined over the past decades. One of the conclusions of their review was that the conceptions of these constructs are not fixed, but continue to move and have become increasingly interchangeable. Nevertheless, in the contemporary broad definition of the concept of self-regulated learning, including metacognitive knowledge and metacognitive regulation of learning (Dinsmore et al., 2008), the origins of the work of Flavell (1976) and Brown (1978) can still be recognised. Flavell studied metacognition in terms of metacognitive knowledge, or in other words, as the knowledge and cognition about cognitive phenomena (Flavell, 1978). He defined metacognition in terms of metacognitive knowledge as “one’s knowledge concerning one’s own cognitive processes and products or anything related to them, e.g., the learning-relevant properties of information or data” (Flavell, 1976, p. 232). Brown viewed metacognition as more dynamic metacognitive regulation, defined as the voluntary control an individual has over his own cognitive processes (Brown & DeLoache, 1978). This second definition of metacognition in terms of self-regulatory mechanisms includes skills such as predicting, checking, monitoring, reality testing, and coordination and control of deliberate attempts to study, learn, or solve problems (Brown, 1978).

14

Chapter 1

These two approaches differ in a sense that metacognitive knowledge describes “…the stable, often fallible, and often late developing information that human thinkers have about their own cognitive processes” (Brown, 1987, pp. 67-78). On the other hand, metacognitive regulation is assumed to be relatively unstable, not necessarily easily to explicate, and relatively age independent (Brown, 1987). According to Brown (1987) these two definitions of metacognition are closely related and attempts to separate them lead to oversimplification. In her view, in order to develop workable theories, it would be necessary to develop them side-by-side with the hope that finally a full understanding of the concept can be attained (Brown, 1987). Nowadays, metacognitive knowledge and metacognitive regulation are still often studied separately and from different research traditions with different methodologies (Lonka, Olkinuora, & Mäkinen, 2004; Pintrich, 2004). In this study, we will try to span both research traditions by studying self-regulated learning including both the more stable knowledge part of self-regulated learning as well as the more dynamic metacognitive regulation of learning (Dinsmore et al., 2008).

1.2.2 Metacognitive regulation Theories describing the dynamic part of self-regulated learning Theories of regulation of learning are based on assumptions about how students learn (Zimmerman & Schunk, 2001) and often use a top-down way of studying students by applying quantitative methods based on psychological constructs and theories (Pintrich, 2004). Regulation of learning is seen as a result of the interaction between biological, developmental, contextual, and individual differences that may hamper or enhance self-regulation at a specific moment (Pintrich, 2004). There are different models of self-regulation of learning with different conceptualisations, but these models share some general assumptions and features (Pintrich, 2000). The first assumption is that learners are active, constructive participants in their learning process. Second, all models assume that learners can potentially control their learning. The third key assumption is that some goal, standard, criterion, or reference value exists that can serve as a gauge against which to assess the operation of the system and then guide regulatory processes. The last assumption is that selfregulation activities mediate between the individual and the context (Pintrich, 2000). These assumptions form the basis of the working definition of selfregulation of learning as an “active, constructive process whereby learners set goals for their learning and attempt to monitor, regulate and control their 15

cognition, motivation, and behaviour, guided and constrained by their goals and contextual features in the environment” (Pintrich, 2000, p. 453). Most models of self-regulation also share the same four phases, sometimes also brought down to three phases (Pintrich, 2000; Winne & Perry, 2000; Zimmerman, 2000). Self-regulation consists of different phases that represent a general time-ordered sequence that individuals go through as they perform a learning task, although different phases can occur simultaneously (Pintrich, 2000). Phase 1 involves planning and goal-setting as well as activation of perceptions and knowledge of the task and context and the self in relation to the task. Phase 2 concerns various monitoring processes that represent metacognitive awareness of different aspects of the self and task or context. Phase 3 concerns efforts to control and regulate different aspects of the self or task and context. Finally, Phase 4 represents various kinds of reactions and reflections on the self and the task or context (Pintrich, 2000). These phases can be directed towards steering one’s cognition, motivation and affect, behaviour, and context. In the past two decades several studies were carried out that contributed to the understanding of the concept of self-regulation by describing individual differences in their regulation of learning (e.g., Schunk & Zimmerman, 1998) and how the development of self-regulation can be fostered (Boekaerts & Niemivirta, 2000; Mevarech & Kramarski, 1997). Furthermore several studies have stressed the importance of self-regulation skills by showing the positive relation between self-regulation and academic learning results (e.g., Pintrich & De Groot, 1990).

Student teachers’ regulation of learning Most of the studies into self-regulation were focused on student learning in academic learning settings. The studies including student teachers or experienced teachers mainly examined how teachers can promote active and self-regulated learning in their students (e.g., Bolhuis & Voeten, 2001; Kramarski & Michalsky, 2009; Niemi, 2002; Perry, 2008; Perry, Phillips, & Dowler, 2004) and they did not include teachers’ regulation of their own learning. Although some studies have been carried out to investigate preservice student teachers’ regulation of learning, these studies only focused on regulation of learning in the academic part of the teacher education programme (Bembenutty, 2007; Chen & Bembenutty, 2005; Corrigan & Taylor, 2004; Hwang & Vrongistinos, 2002; Järvenoja & Järvelä, 2007; Meeus, Van 16

Chapter 1

Petegem, & Meijer, 2008; Orange, 1999; Ross, Salisbury-Glennon, Guarino, Reed, & Marshall, 2003). The participating student teachers in these studies did not differ from regular college students in a sense that they followed a course at the university and were not involved in teaching in practice at the moment of investigation. Although studies which describe student teachers’ regulation of learning when involved in teaching practice hardly exist, some studies have been carried out to explore to what extent experienced teachers engage in regulation of learning in their authentic work context. Both studies of Kreber, Castleden, Erfani, and Wright (2005) and Van Eekelen, Boshuizen, and Vermunt (2005) showed that academic staff varies in the degree of engagement in self-regulation activities, in particular, differences were found in the extent to which these teachers set explicit goals for themselves. In the study of Van Eekelen and colleagues (2005) less than one third of experienced teachers’ learning experiences could be categorised as self-regulated learning in a sense that they were planned beforehand. The study of Lin, Schwartz, and Hatano (2005) concluded that conventional metacognitive interventions are not applicable for teachers, since the assumptions underlying these interventions, especially when it comes to goal setting, do not match the complex teaching situations. Unlike the contexts of most metacognitive interventions, teaching situations do not have well-defined problems, are not stable learning environments, and lack shared values and goals (Lin et al., 2005). Comparable to the conclusions of Lin and colleagues (2005), Randi (2004) also stated that teachers may draw on different self-regulatory skills than their students do. Hence, a different conceptual framework might be needed to give a good description of the quality of student teachers’ regulation of learning.

1.2.3 Metacognitive knowledge Theories describing the stable part of self-regulated learning Historically, metacognitive knowledge was focused on specific factual knowledge students had about their learning (Flavell, 1976). Nowadays, the more stable part of self-regulated learning is also studied as more integral conceptions of learning. This perspective is called the student approaches to learning perspective and also started in the seventies when Marton and Säljö (1976) described the qualitative different ways of processing information among university students reading an academic article as deep and surface levels of processing. A few years later, Säljö (1979) used the same bottom-up 17

phenomenographic methodology to describe students’ conceptions of learning. He found a hierarchy of five learning conceptions in which learning was conceptualised by students as: (a) increasing one’s knowledge, (b) memorising and reproducing, (c) applying, (d) understanding, (e) seeing something in a different way. This taxonomy was confirmed by a study of Marton, Dall’Alba, and Beaty (1993), who also added a sixth conception: (f) changing as a person. The first three conceptions describe quantitative views on learning in which learning is seen as reproduction of material which reflects a surface approach, and the latter three are more qualitative views and focus on the role of meaning in learning, reflecting a deep approach (Biggs, 1979; Marton et al., 1993; Säljö, 1979). Reproduction-oriented learning and meaning-oriented learning have remained the main two dimensions for describing individual differences in students’ learning conceptions (Marton & Säljö, 1976), learning approaches (Biggs, 1988; Entwistle, 1988), learning styles (Vermunt, 1996), learning patterns (Vermunt & Vermetten, 2004), study orchestrations (Meyer, 1991), and learning orientations (Lonka & Lindblom-Ylänne, 1996). Nowadays, inventories are available to measure several aspects of learning with different scales, resulting in factors described as orientations (Entwistle & Ramsden, 1983; Lonka et al., 2004) or styles (Vermunt & Van Rijswijk, 1988), which are regarded as more stable across various situations than approaches in their original definition (Lonka et al., 2004). Among university students, Vermunt (1998) found undirected, reproduction-directed, meaning-directed, and application-directed learning styles. In these styles or orientations, besides learning conceptions and cognitive strategies, another dimension in student learning was also added, namely the students’ regulation activities or metacognitive strategies. Although this regulation dimension is clearly a separate dimension on the conceptual level (Entwistle & McCune, 2004), often empirical relations have been found with approaches to learning: self-regulation is most often related to the deep approach and external regulation to the surface approach (e.g., Lonka & Lindblom-Ylänne, 1996; Rozendaal, Minnaert, & Boekaerts, 2003). This can be understood, since searching for meaning involves monitoring and reshaping one’s thought, while the surface approach is merely reactive (Biggs, 1988). Studies using these inventories were also able to confirm the expected relation between a meaningdirected learning and positive learning outcomes (Lindblom-Ylänne & Lonka, 1999; Vermunt & Vermetten, 2004).

18

Chapter 1

Although the stability of learning orientation or styles is expected to be quite high, these are not unchangeable traits of individuals (Vermunt, 1998). Several developmental phenomena have been observed, for example, more experienced learners develop a more coherent integration of their learning conceptions, motives, and strategies (Vermunt & Verloop, 2000). On the other hand, students’ conceptions, motives, and strategies also become more differentiated as learners have greater experience with different domains (Buehl & Alexander, 2006). Furthermore, the application-directed learning style develops relatively late and emerges not always from analyses as a distinct learning style (Vermunt & Vermetten, 2004). These changes in learning orientation are often provoked by changes in the learning environment (e.g., Tynjälä, 1997), but this often co-occurs with temporary dissonant patterns, in which learning conceptions and strategies do not match (Lindblom-Ylänne & Lonka, 2000; Vermunt & Verloop, 2000).

Student teachers’ learning orientations Although some studies from the student approaches to learning (SAL) perspective have been done in the context of pre-service teacher education, most of these studies focused on student teachers’ learning conceptions and orientations considering the academic part of the teacher education programme (e.g., Brownlee, Purdie, & Boulton-Lewis, 2003; Donche & Van Petegem, 2009; Van Petegem, Donche, & Vanhoof, 2005). Kremer-Hayon and Tillema (1999) examined the general attitude of student teachers and their teacher educators towards SRL. Their study showed that student teachers are more positive towards SRL and have higher expectations about their own selfregulative competences than their teacher educators. Oosterheert and Vermunt (2001) studied extensively individual differences in learning to teach in several Dutch dual pre-service teacher education programmes. Their interview data discerned five different orientations towards learning to teach. One orientation described student teachers holding tenaciously to their own field experience and scarcely being able to report on self-regulation (survival-oriented). The other learning orientations differed in whether they were meaning- or reproduction-oriented and how they approached their problems (open or independent versus closed or more dependent from others) (Oosterheert & Vermunt, 2001). In a follow-up study, an inventory was developed (Inventory Learning to Teach Process, ILTP), to study these orientations on a larger scale (Oosterheert, Vermunt, & Denessen, 2002). In this second study the existence 19

of the initial orientations was confirmed, except for the open reproductiondirected orientation. Oosterheert (2001) expected that student teachers would develop from survival-oriented or reproduction-oriented toward dependent meaning-oriented learning, ending in independent meaning-oriented learning. Furthermore, she hypothesised that different orientations would also lead to qualitatively different learning outcomes.

1.3 Conceptualisation of self-regulated learning for the present studies To get an integrated view of both parts of self-regulated learning, it is in our opinion necessary to study both the more stable knowledge part of selfregulated learning and the more dynamic metacognitive regulation of learning simultaneously. Historically, the stable knowledge part of SRL was limited to the specific facts students had about learning and memory, but in the last two decades a more coherent picture of students’ conceptions of learning and associated concepts have been studied (Vermunt, 1996). For the context of preservice teacher education, Oosterheert and Vermunt (2001) used the concept of learning orientations to denote typical combinations of learning conceptions and preferences for learning- and regulation activities. The concept of learning orientation can be studied on three levels of context: general orientations, course-specific, and situational orientations (Lonka et al., 2004). If learning orientations are measured without a specific learning situation in mind, they refer to the learner’s general disposition to learning (Vermetten, Lodewijks, & Vermunt, 1999). By using the concept of learning orientations for studying the more stable part of self-regulated learning, we expect to get a detailed description of student teachers’ general dispositions to learning. Student teachers’ (dynamic) regulation of learning is operationalised for our studies as the combination of regulation activities they perform in concrete learning experiences. Since the term metacognitive regulation is not often used any more, the dynamic part of self-regulated learning is in literature also referred to as self-regulation, self-regulation of learning or self-regulated learning, which can be quite confusing. In this dissertation we will reserve the concept of self-regulated learning for referring to the combination of the dynamic and the stable part of the concept. Self-regulation, student teachers’ regulation of learning, and self-regulation of learning will be used to refer to the dynamic part of the concept. To study the stable part of SRL, the concept of learning orientations 20

Chapter 1

will be used. In Figure 1.1, a schematic overview is given of the concepts used in this dissertation.

self-regulated learning dynamic part

conceptualisation for our studies

definition

self-regulation of learning

stable part

learning orientation

(also referred to as: self-regulation, regulation of learning, student teachers’ regulation of learning)

student teachers’ combination of regulation activities as performed in concrete learning experiences

student teachers’ combination of their general learning conceptions and preferences for certain learning and regulation activities

Figure 1.1. Overview of the concepts used in this dissertation.

1.4 Focus and overview of this dissertation From the review of the literature as described above, multiple conclusions can be drawn. First, research into both parts of SRL has provided valuable insights into students’ self-regulated learning. In addition, self-regulation of learning is also assumed for student teachers to be a necessary skill to be able to maximise learning in a dual learning programme, as well as an important aspect of becoming and being an adaptive expert as a teacher. Therefore, one of the main purposes of teacher education programmes should be to support the development of these self-regulated learning skills to prepare them for lifelong learning. The dynamic part of SRL, however, has primarily been studied in the context of academic learning. This means that we know a lot about how students regulate a very specific form of learning, namely studying and accomplishing tasks which are often designed by others. Detailed descriptions of how student teachers regulate their learning in a dual learning programme in which student teachers design their own learning experiences are lacking. Furthermore, student teachers differ in their orientations towards learning to teach. Multiple studies have shown that an active and independent meaningoriented way of learning is regarded as being preferable for becoming an expert teacher. However, it is not clear how or to what extent student teachers develop towards this way of learning throughout the programme. Therefore, 21

the main research question of this thesis is: What is the nature and development of student teachers’ self-regulated learning in a dual learning programme? To answer the main research question regarding the nature and development of student teachers’ self-regulated learning, two studies were designed. The study of Oosterheert and Vermunt (2001) already provided a rich conceptual framework describing the nature of student teachers’ learning orientations. Therefore, the first step was to obtain insights into student teachers’ regulation activities which they used across multiple learning experiences in a dual teacher education programme. A small-scale qualitative cross-sectional study was carried out in 2005/2006 in which 28 student teachers were included. In Chapter 2, the variation in and underlying structure of these regulation activities will be explored, aimed at answering the following research question: (1) What is the nature of student teachers’ regulation of learning across different learning experiences in the two contexts of a dual learning programme? In this chapter, the unit of analysis is the learning experience. Chapter 3 will describe differences in self-regulated learning on the level of the individual student teacher by looking at the patterns of student teachers’ regulation of multiple learning experiences, for which the same data have been used as in Chapter 2. In this small-scale study, also student teachers’ more stable part of SRL was measured by means of interviews for which the conceptual framework of Oosterheert and Vermunt (2001) was used. In the interviews, however, no detailed information was gathered about the specific preferences for learning and regulation activities, therefore only the student teachers’ general learning conceptions were categorised. The combination of learning conceptions together with analyses of individual patterns of regulation of learning provided insights into the relation between the two parts of the concept of self-regulated learning. To conclude, in Chapter 3, the following research question will be answered: (2) What are differences in student teachers’ regulation patterns and learning conceptions in a dual learning programme and how are these related to each other? In the small-scale study, an open question log was used to measure student teachers’ regulation activities in a dual learning programme. In order to collect longitudinal data in a larger sample, a less labour-intensive way than this open question log was needed for the student teachers as well as the researchers to collect data from multiple learning experiences. Therefore, the next step was to develop a less labour-intensive instrument that was able to measure regulation of learning in a valid and reliable way. In Chapter 4 the 22

Chapter 1

development and evaluation of this instrument will be described, based on the following research question: (3) To what extent is a structured multiple event question log a valid and reliable instrument for measuring and discriminating different qualities of student teachers’ regulation of learning in a dual learning programme? The final longitudinal larger-scale study was conducted to confirm the framework as developed in the small-scale study and moreover, to describe the development in both parts of student teachers’ self-regulated learning. For this study, in which 81 student teachers from two programme tracks were included, the Structured Learning Report as described in Chapter 4 and the ILTP (Oosterheert et al., 2002) was used to study the combination of student teachers’ learning conceptions and preferences for learning and regulation activities. These programme tracks included a track followed by student teachers with a paid job and an intern-track (see also section 1.5). In Chapter 5 the findings of this study are presented aimed at answering the following research question: (4) How do student teachers’ regulation of learning and learning orientations change throughout a one-year dual learning programme and what are differences between the two programme tracks? A schematic overview of this dissertation, including the foci of the chapters is displayed in Figure 1.2. self-regulated learning dynamic part

self-regulation of learning nature

learning orientation

Question 1 (Chapter 2)

individual differences measurement

stable part

Question 2 (Chapter 3)* Question 3 (Chapter 4)

development

Question 4 (Chapter 5)

data from the small-scale study data from the larger-scale study

General discussion (Ch. 6)

not included in the study * only the learning conception part of the concept of learning orientation was included in this study

Figure 1.2. Overview of this dissertation1. The studies reported in Chapters 2 through 5 have been submitted to international journals. Therefore, some overlap in texts between chapters may occur.

1

23

1.5 The context for the present studies The nature and development of student teachers’ SRL has been studied in a postgraduate one-year teacher education programme in The Netherlands, preparing for subject-teaching in upper secondary school. The way universitybased teacher education for upper secondary schools is organised in The Netherlands is comparable with the Finnish model for subject-specific teacher education in which students first get their subject education of 240 credits, followed by one year of teacher education including learning from theory and practice (Tryggvason, 2009). This postgraduate programme is a dual learning programme, meaning that the student teachers learn in parallel in practice and at the university. Because of the shortage of teachers in The Netherlands, many student teachers already have a job or have applied for a job when starting the teacher education programme. Therefore, the teacher education has a job-track as well as an intern-track. Student teachers who already have a job or applied successfully for a job follow the job-track and start from the first day as a teacher at a secondary school. The majority of these student teachers have some teaching experience before starting this job-track. However, due to the shortage of teachers in The Netherlands, especially for the science subjects, also student teachers without prior experience get a job and follow the job-track as a novice. The other student teachers are following the internship track, in which they are more gradually exposed to the teaching profession. Their first internship is in triads: they observe their peer student teachers and experienced teachers, having their first teaching experiences and reflect on these together. In the second semester, the interns change schools and do the internship alone while having more responsibilities as a teacher, including an increase in teaching lessons. The content of the academic part of the programme is the same for both tracks. All student teachers attend weekly general and subject-specific classes at the university organised by teacher educators. In these classes, they also meet other student teachers to discuss and learn about several aspects of the teaching profession. The student teachers designing their personal curriculum, based on their prior experiences and concerns, which they discuss with their teacher educator. Based on this personal development plan, they choose on which topics they want to specialise and which extra theoretical lectures they want to attend. The end terms of all teacher education programmes in The Netherlands are based on the teaching competences defined by The Association for the 24

Chapter 1

Professional Quality of Teachers, (SBL). This association formulated required competences for teachers to promote a standard quality of education. SRL is an important part of the competence “reflection and development”. In this competence it is described that a teacher in secondary education “must permanently work on his personal and professional development. That is his/her responsibility, and in order to take this responsibility the teacher must possess competence for reflection and development” (SBL, 2004, p. 20). Student teachers in the programme under study keep a portfolio in which they describe their self-evaluations and their personal development plan to direct their learning processes with respect to the different competences. In this personal development plan they also have to reflect on how they take responsibility for their own development. Furthermore, student teachers have to provide evidence of their competences by including the products they have made and feedback they received, for example lesson plans, observations, teaching videos, and assignments. Their assessment at the end of the programme is based on their portfolio and a final interview; no grades or other quantitative measurements are used.

25

academic forethought variance asked

stimulate

education workplace

urgent explained situational emerging multiple

self-regulation

factor

instructed peers

beliefs

low-scoring

variety

source whereby

content

unintentional

mca

coded involve

experienced explicit

related

characteristics

models

namely

self-efficacy

turned

behaviour

superficial

action

lack

intentional covering

distinction

implicit

kappa

aspects

contextual

monitoring

described

domain

indicators

results

focus become

external underlying orientation research deliberative

categorisation

pupils

influence

fragments following

further

student teachers'

focused

appeared

qualitative personal cases weeks

kinds passive programme framework

specific

corno

notions

participate

classroom frequency

retrospective

reports

interpreted

combination

consciously

vermunt

different changes outcome

strategy

learning reports

designing

obtain

description

ends

regard

iterative

activities

knowledge

experiences development goal-setting

pintrich

opportunities

reflection

reactive

reasons

institute high-scoring structure analysis included similarities task variation data conducted

concerning

called

students

higher

categories

variables

primarily learners leeuw achievement

terms

srl

doing

information

dual

one-third

actually components

suggests

understanding relevant

identified argument spontaneous various

analytic

learning

classified

limitations

findings

dimension

university

willing

previous

self-chosen

versus referred evaluation

planned

conceptualised

goals

expectations

colleagues

theories teaching

context

generally

exploring

unplanned

post-graduate procedure

concept

prospective

deliberate

works

phase

regulation

principal

steering

differentiation

conclusion

represents

teacher

varied weekly measurement

2 The nature of student teachers’ regulation of learning in a dual learning programme2 Abstract Self-regulated learning has mainly been conceptualised to involve learning within academic settings. Student teachers who learn in dual learning programmes, however, also need to regulate their learning as a teacher. Hence, there is an urgent need to extend self-regulated learning theories to the domain of learning to teach and to obtain scientific knowledge on the nature of student teachers’ regulation of learning to enable support of these processes in teacher education. Therefore, this study was aimed at exploring the nature of student teachers’ regulation of learning in a dual learning programme. Twenty-eight students from a post-graduate academic teacher education institute participated in this study. For the measurement of student teachers’ regulation activities an open question log, called Learning Report, was developed. Content analysis of 133 Learning Reports resulted in the identification and description of the variety and frequency of student teachers’ regulation activities. The relations among the regulation activities could be described by an underlying structure of two dimensions: passive versus active regulation of learning and prospective versus retrospective regulation of learning. Active regulation dominated in practice schools, passive regulation at the university. It is argued that for the context of dual learning programmes a different conceptualisation of SRL is needed, focusing less on setting initial learning goals and more on retrospective aspects of self-regulated learning. Building blocks for such a conceptualisation are offered.

This chapter has been submitted in adapted form as: Endedijk, M., Vermunt, J.D., Verloop, N., & Brekelmans, M. (under revision). The nature of student teachers’ regulation of learning in a dual learning environment. 2

27

2.1 Introduction To date, research into self-regulated learning (SRL) has focused primarily on student learning in traditional instructional settings. These studies have provided insights into processes and models of SRL, and useful applications of theory for school practice (Schunk, 2005). Not only students, however, draw on self-regulation strategies to cope with situational demands in the classroom; the same is true of teachers (Randi, 2004). Self-regulation skills are necessary for teachers in order to become expert teachers instead of only experienced non-expert teachers (Kreber, Castleden, Erfani, & Wright, 2005). Also, from the perspective of lifelong learning (Cornford, 2002) and because teachers as experts in learning have to act as role models for their students, it is very important that teachers also learn to regulate their own professional learning (Hagger, Burn, Mutton, & Brindley, 2008; Kramarski & Michalsky, 2009). Nevertheless, SRL has mainly been conceptualised to involve student regulation within academic learning settings (Kaplan, 2008).

Although in the past most teacher education institutions offered university-based curricula aimed at applying theories in practice with little opportunity for self-regulated learning (Kremer-Hayon & Tillema, 1999), teacher education programmes have increasingly been organised as dual learning programmes. These dual learning programmes, in which student teachers combine learning from studying at the university and learning from practice at schools, require a high degree of SRL from student teachers. They also have to integrate knowledge gained from teaching experiences and university courses, and often have to self-evaluate their competences, clarify their learning needs, formulate personal development plans, reflect on their learning, and document their learning progress in a portfolio. It is unclear, however, whether student teachers are sufficiently able to regulate their learning processes to this extent, and how educational institutes can stimulate the development of SRL. Hence, there is an urgent need to extend SRL theories to the domain of teacher learning and to obtain scientific knowledge on the nature of student teachers’ regulation of learning in order to understand and support these processes. Therefore, this study was aimed at exploring the nature of student teachers’ regulation of learning in a dual learning programme.

28

Chapter 2

2.2 Theoretical framework SRL has generally been conceptualised with a focus on regulation within academic learning settings (Kaplan, 2008) and has been studied in classrooms as well as laboratories (Dinsmore, Alexander, & Loughlin, 2008). Selfregulation is an umbrella term for a variety of activities. In most theories, selfregulation of learning is defined as an “active, constructive process whereby learners set goals for their learning and attempt to monitor, regulate and control their cognition, motivation, and behaviour, guided and constrained by their goals and contextual features in the environment” (Pintrich, 2000, p. 453). This process consists of different phases that represent a general time-ordered sequence that individuals go through as they perform a learning task, although different phases can occur simultaneously (Pintrich, 2000). Phase 1 involves planning and goal-setting as well as activation of perceptions and knowledge of the task and context and the self in relation to the task. Phase 2 concerns various monitoring processes that represent metacognitive awareness of different aspects of the self and task or context. Phase 3 concerns efforts to control and regulate different aspects of the self or task and context. Finally, Phase 4 represents various kinds of reactions and reflections on the self and the task or context (Pintrich, 2000). Although SRL is by most definitions restricted to students in academic contexts, the concept is valuable as well for teachers in taking charge of their own learning (Randi, 2004). Randi (2004) suggests that teachers may need different self-regulatory skills than their students. This leads to the intriguing question which regulation activities student teachers use in a dual learning programme in which they are students at the university and real teachers in schools at the same time. During learning in practice, for example, most of the time there is no teacher educator present to regulate student teachers’ learning externally. Furthermore, their primary task is to focus on their students’ learning, but they are also in charge of their own learning processes and have to recognise learning opportunities while teaching at the same time. In a dual learning programme, student teachers can learn from many different sources of information, and daily practice at the workplace has a big influence on their learning processes. Learning at the workplace is less intentional and planned, and more contextual and collaborative, than learning in traditional learning settings (Hager, 2004; Tynjälä, 2008). Besides, teacher learning does not have preset objectives or easily identifiable outcomes (Hodkinson & Hodkinson, 2005). In conclusion, the nature of teachers’ learning differs in many aspects 29

from the learning processes of their students, which also might influence the nature of their regulation of learning. Until now, very few empirical studies investigated how student teachers or experienced teachers themselves engage in self-regulated learning. The study of Van Eekelen, Boshuizen, and Vermunt (2005) has shown that in the context of higher education only one-third of the learning of experienced teachers is self-regulated in some way. Furthermore, for these higher education teachers it is often hard to see themselves as learners about teaching (Kreber et al., 2005). The few studies that have been conducted in pre-service education have detected student teachers’ conceptions of or attitudes towards SRL (Kremer-Hayon & Tillema, 1999), or have given suggestions for improving teacher education (Lin, Schwartz, & Hatano, 2005). Some have focused on the preferences of student teachers for a certain learning styles. The results of these studies have shown that student teachers claim to rely minimally on selfregulation strategies (Donche, Vanhoof, & Van Petegem, 2003). Oosterheert and Vermunt (2001) found five different orientations towards learning to teach. One orientation described student teachers holding tenaciously to their own field experience and scarcely able to report on self-regulation. The other learning orientations differed in terms of self-regulation versus external regulation, and on the dimension meaning-oriented versus reproductionoriented learning (Oosterheert & Vermunt, 2001). In conclusion, most of the research that has been done in the context of learning to teach focused primarily on the orientations, attitudes or conceptions of student teachers towards SRL. Student teachers generally report that they do not rely much on self-regulation strategies; however, to our knowledge no research has been done on the nature of the regulation activities student teachers actually undertake in different parts of a dual learning programme. Surprisingly, within the domain of student learning also only a few studies focused on the qualitative differences in regulation activities between students, probably because this type of investigation is time-consuming (Winters, Greene, & Costich, 2008). Models of SRL have therefore been based more on analytic than on empirical activities (Winne, 2005). According to Kaplan (2008), in the whole field of SRL research there is an urgent need to conceptualise self-regulated action in various dimensions in order to get a more detailed understanding of the nature of SRL. This dimensional framework could then provide insight into the qualitative differences in students’ regulatory action during their development (Kaplan, 2008). Because 30

Chapter 2

of the lack of understanding of the nature of self-regulation of learning in a dual learning programme such as teacher education, this study was aimed at exploring the variation and dimensions in student teachers’ regulation activities during different learning experiences in the context of the dual learning programme. More specifically, we aimed to answer the following research questions: 1. What regulation activities do student teachers undertake in a dual learning programme? 2. Which underlying dimensions can be found to describe the relation between these regulation activities? 3. Are there differences between the two different contexts of the dual learning programme (university and practice schools) with regard to the nature of student teachers’ regulation activities? This study was conducted in a post-graduate teacher education programme; students first obtain their master’s degree in a specific subject area and then enrol on a one-year programme to obtain their teaching degree for upper secondary education. During the programme, student teachers attend weekly classes at the university while also doing teaching practice at schools or holding a paid job as a teacher. They have much freedom in designing their personal curriculum, which is based on their prior experiences and concerns. They keep a portfolio in which they make self-evaluations and a personal development plan to direct their learning processes. In this personal development plan, they also reflect on their own responsibility for their learning processes.

2.3 Method 2.3.1 Participants Twenty-eight student teachers from one Dutch post-graduate teacher education programme volunteered to participate in this study. The student teachers were selected with the aim of capturing as much variation as possible in the following variables: teaching experience, supervising teacher educator, and school subject. Seven student teachers who were not willing to participate (primarily because of lack of time) were replaced by other student teachers with the same characteristics in terms of these variables. Nine of the students were male, nineteen were female. 31

2.3.2 Instrumentation For the measurement of qualitative differences in student teachers’ regulation activities an open question log, called Learning Report, was developed. In this instrument, student teachers were asked to describe a self-chosen learning experience, stimulated by eight open questions about different regulation activities. The whole instrument consisted of six Learning Reports. The questions in the Learning Report were based on Pintrich’s (2000) conceptual model, which represents a blueprint for the development of new instruments for measuring SRL. For every phase of SRL two or three questions were included in the Learning Report: for the planning and goal-setting phase questions were inserted about student teachers’ goal orientation (Question 2), notions of self-efficacy (Question 3), learning strategy (Question 4) and forethought (Question 8). The questions concerning the monitoring and control phase described their control of the learning strategy (Question 5) and monitoring of the learning results (Question 6). The questions concerning the reflection phase were focused on the learning outcome (Question 1) and evaluation of the learning experience (Question 7). The description of the questions can also be found in Table 2.1 in the Results section. Phases 2 (monitoring) and 3 (control/regulation) were taken together, because between these two processes not much differentiation has been found in self-report instruments (Pintrich, 2000). Besides these eight regulation questions, we also asked the student teachers to describe the learning context of their learning experiences to increase the interpretability of the Learning Reports. The instrument was tested in a small pilot study with three student teachers who completed six Learning Reports each; the results led to some small corrections in the phrasing and order of the questions. It appeared that the student teachers had a desire to start their Learning Report by reflecting on the learning experience; therefore we made the reflection question the first question in the Learning Report. Furthermore, the student teachers reported that they had no problems recalling different aspects of regulation some time after the learning experience, except for the question concerning their original forethought of that learning experience. Therefore, we decided not to let them recall their original forethought, but to describe in the last question of the Learning Report their forethought for a new future learning experience.

32

Chapter 2

2.3.3 Procedure Following the first contact by phone or e-mail, the student teachers received instructions about the Learning Reports orally during an individual meeting with the researcher, and afterwards also on paper. They were asked to choose a learning experience from the previous week; this could be any kind of experience that was part of their development as teachers. Because of the diversity of student teachers’ learning processes, as described previously, they were stimulated to describe different kinds of learning experiences in the six Learning Reports: at least two learning experiences that had taken place in the context of the teacher education institute, and at least one from the context of the practice school. Furthermore, they were instructed that they could report planned as well as spontaneous learning experiences, and both successful and unsuccessful learning experiences. They were told that in the case of an unplanned learning experience, some questions were not relevant so they had to answer these questions with not relevant. Data were gathered during a period of six consecutive weeks. To stimulate student teachers to complete the reports and send them back weekly, we sent them a weekly reminder by e-mail, and they were phoned when they did not respond within two weeks. In total, 134 Learning Reports were collected; eighteen student teachers completed all six Learning Reports as required, and four student teachers withdrew after four or five Learning Reports. Two student teachers did not send in more than three Learning Reports, and three others did not complete a single report. One student teacher only completed one Learning Report, which was excluded from analysis, because the reported experience was not related to learning to teach. The reasons student teachers gave for stopping before completing all six reports were mostly illness, quitting teacher training, or lack of time.

2.3.4 Data analysis The data were analysed in two phases. The first phase was iterative. The Learning Reports were read through repeatedly to produce a set of categories covering the variation in self-regulation in the reported learning experiences. Our first step of this phenomenographically informed content analysis was reading all the Learning Reports several times in order to become familiar with the data. To reach a set of distinctive categories, a process of continuous comparison of the quotes with the emerging categories was needed. For this process, the program ATLAS.ti 5.0 was used (Muhr, 1997), which made it 33

possible to have the complete Learning Report visible during the categorisation of a fragment in order to maintain context-boundness. During this process eight variables were established, each representing a different aspect of the regulation process and one of the questions of the Learning Report, around which a set of categories could be constructed. After every iteration, colleague researchers critically examined the preliminary set of categories for each variable. Five iterations were needed to reach a stable set of categories covering the qualitatively different descriptions of self-regulated learning in the Learning Reports in as few categories as possible. All text fragments of the Learning Report were coded in terms of the set of categories developed. From the total of 1197 fragments that were coded, 10% were also coded by an independent second researcher to calculate inter-rater reliability. In the first round, an overall Cohen’s Kappa of .86 was reached. Although this appeared satisfactory, it emerged that the Cohen’s Kappa for the different separate variables varied from .70 to 1.00, except for one variable (Forethought) for which a Cohen’s Kappa of only .36 was reached. For this variable the code book was made more specific and another 10% of the data were coded to test the changes. This led to a new Kappa for this specific variable of .80 and a new overall Cohen’s Kappa of .90. In the second phase, multiple correspondence analysis (MCA) was used, also referred to as homogeneity analysis (Mair & De Leeuw, 2008) which is a form of nonlinear principal component analysis (De Leeuw, 2005). This technique makes it possible to discover relationships between categorical data and the underlying structure in the data and is especially useful when there are many variables and categories (Mair & De Leeuw, 2008). MCA is a nonparametric factor analytical procedure and like parametric factor analysis it orders a number of possibly related variables along a small number of principal components which explain (part of) the variance of the original variables. In MCA these principal components are called dimensions and, after the number of dimensions has been chosen, the outcome of the MCA shows the position of the different values of the variables (the categories) on these dimensions. Learning Reports characterised by the same categories in the different variables are plotted close together and Learning Reports that do not resemble each other are plotted far apart (Funnell, Bryer, & Grimbeek, 2004). In conclusion, this technique quantifies and visualises which categories and which Learning Reports can be considered as similar or related by putting them close to each other in the solution. By describing and analysing these 34

Chapter 2

similarities and dissimilarities, the dimensions and underlying structure in the data can be interpreted.

2.3 Results 2.3.1 Student teachers’ regulation activities in a dual learning programme Content analysis of the 133 Learning Reports showed that the variation within each of the eight aspects of regulation could be described by five to seven categories. To answer the first research question, a global description of the range of categories will be given below separately for every variable. An overview of the total set of variables and categories, including their full descriptions, illustrative examples from the data and the observed frequencies, can be found in Table 2.1. Learning goal orientation (variable A in Table 2.1) described the reason why student teachers learned what they learned. The categorisation is based on the answer to Question 2: did you plan to learn this and if so, why did you want to learn this? In total, 83 learning experiences were said to be intentional, and 50 were reported as unintentional. Within the intentional learning experiences, the reported reasons covered two main aspects: a description of the unsatisfactory current situation which they wanted to change and the learning goal a student teacher aimed to reach. Sometimes only one aspect was mentioned as an argument for learning, but aspects were also mentioned in combination. Self-efficacy notions (variable B in Table 2.1) concerned student teachers’ ideas with respect to their expectations of succeeding in a learning experience. In only nine cases did student teachers expect not to succeed. Independently of whether they expected to succeed or not, however, they used different kinds of information in order to form their expectations, which were categorised for this variable. Often student teachers reported that this was about trust in their own capacities, but they also based their expectations on their previous experiences with the learning strategy, learning context or learning content.

35

36

Category description

Example from the data (summarised)

No answer had been given.

No answer …

I have been struggling for some while now with my reflection skills, I want to change something in that

I was well prepared and discussed this also with a colleague I hoped that I would learn something during the lecture

The ST expected a certain amount of success because of her previous experience with the learning context.

The ST expected a certain amount of success because of her confidence in her own qualities or effort.

The ST only hoped to succeed, but did not mention a reason for this.

Experience with learning context

Own qualities/efforts

Hope without argumentation

The pupils I was going to teach are always very cooperative

Lectures always help me to reflect on myself as a teacher

The ST expected a certain amount of success because of her previous experience with the learning strategy.

Experience with learning strategy

We had discussed multiple intelligence already

The ST expected a certain amount of success because of her previous experience with the learning object.

Experience with learning object

B Self-efficacy notions: ‘Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this?’ (Question 3)

(question was not relevant, because of an unplanned learning experience)

ST described the direction in which she wanted to grow, including an explicit goal, but without describing the starting problem explicitly.

Direction of growth with judgement of current situation

(not

I wanted to learn more about collaborative learning

ST described the direction in which she wanted to grow, including an explicit goal, but without describing the starting problem explicitly.

Direction of growth with learning goal

relevant) 3

Until now, I did not succeed in being strict, so that is what I want to reach

Both a judgement of current situation and a new explicit goal had been described.

Judgement of current situation & learning goal

I want to know that all my children learn something from my lesson

ST described a new goal, but there was no information about the relation with the current situation.

Learning goal

I was unsatisfied about what happened

ST described the current situation, but no new goal or direction of growth had been described.

Judgement of current situation

A Learning goal orientation: ‘Did you plan to learn this, and if so, why did you want to learn this?’(Question 2)

Category

5

22

14

12

9

-0.51

0.63

0.76

-1.10

0.18

-0.88

2 50

0.26

-0.25

-0.13

0.14

0.17

Dim 1

16

12

26

22

5

N

0.27

0.35

0.78

0.30

-0.04

-0.05

0.43

1.03

0.27

0.25

-0.29

Dim 2

Table 2.1 Variables and categories of student teachers’ regulation activities, including their description, an example from the data, observed frequencies (N), and category loadings on Dimension 1 (Dim 1) and 2 (Dim 2) of the multiple correspondence analysis.1, 2

I discussed this problem with my mentor I learned this by listening and watching a DVD First the teachers explained it to me and then I applied it to my own lessons

ST learned by getting feedback or information from others or more indirectly by observing others.

ST learned by listening or reading or other ways of processing information.

ST learned by applying information or theories to her own practice.

No answer had been given.

Interacting or getting feedback

Processing information

Applying theory to practice

No answer

The ST did not choose herself a specific learning strategy, this had been part of an instruction.

ST explained that is had not been a conscious choice to learn in this specific way.

No answer had been given.

(question was not relevant, because of an unplanned learning experience)

Part of an instruction

No conscious choice

No answer

(not relevant) 3,4

Reflection on own performance

After reflection on the experience, by analysing her role in the event, the ST knew what she had learned.

E Monitoring of the learning results: ‘How did you realise that you had learned something?’(Question 6)

I prefer processing information by reflecting on it or discussing it with others

ST gave an argument for a way of learning.

Argument for a learning strategy

By looking back I realised what went well and what was needing some improvement



It was not a conscious choice to learn it that way

We had to learn this via an authentic case

This pedagogy is clear for my pupils and it has immediately effect

ST gave an argument for a way of teaching, so why the ST had adapted a certain pedagogy in the classroom.

Argument for a way of teaching

D Learning strategy control: ‘Why did you learn it in this way?’(Question 5)

After evaluating the problem, I realised that I need to define my core problem in order to develop myself further.

ST learned by evaluating a (teaching) experience or reflecting upon it.

Reflecting or evaluating



I learned this by just trying and seeing what happened

ST leaned by doing or experiencing things.



Learning by doing

C Learning strategy: ‘How did you learn this?’(Question 4)

No answer had been given.

(question was not relevant, because of an unplanned learning experience)

No answer

(not relevant) 3

Table 2.1 (Continued) 21

20

48

60

4

5

12

4

17

9

24

14

15

54

50

0.20

0.05

-0.07

-2.09

0.34

0.62

-0.25

0.24

-1.77

-0.03

0.72

0.63

-0.50

-0.01

0.31

0.95

0.43

0.29

0.74

-1.52

0.99

0.12

-0.89

-0.06

0.14

0.32

Chapter 2

37

38

ST became aware that she had learned something, because she had had new information.

No answer had been given.

Novelty of information

No answer

Structuring homework assignments

The object of learning was described as a specific teaching behaviour.

The ST described examples of knowing why in terms of when, why and under what conditions what kind of teaching is important or good to do.

An experience has been described, but it was not clear what had been learned from it.

Specific teaching practice

Knowing why

Description of an experience

It gave me good insights, but next time I have to prepare myself better for the lecture Actually it would have been better if I could have kicked the student out of my classroom

ST evaluated the learning process, what could have been improved or what she was satisfied with.

ST evaluated primarily her own behaviour in the experience.

Evaluation of learning strategy

Evaluation of learning context or own behaviour

G Self-evaluation of the learning process: ‘If you look back, about which aspects are you satisfied and what would you do differently next time?’ (Question 7)

We talked about motivation

If you can show the relevance of your lesson, pupils will listen, be quieter and pay more attention

I learned to trust in my own skills

How you can deal with speaking skills in the classroom

The learning object was described as knowledge about teaching behaviour.

The learning object was described as a change in the ST’s identity or awareness of her learning process.

That performance anxiety is related to perfectionism

Knowing about myself

The learning object had been described as knowledge or information, as a learned fact.

Knowing that

Never start a lesson with a negative feeling towards pupils



Knowing how

ST described what she had learned on the level of a rule of thumb, a strict rule that always has to be followed.

Rule of thumb

F Reflection on the learning object: ‘What did you learn?’(Question 1)

After communicating with others, I realised that not only external factors have influence

After reflecting on information from others, the ST realised that she had learned something.

Reflection on information of others After the meeting, I had new and concrete information

My pupils were acting more relaxed and were motivated to cooperate

Getting information from others, sometimes only by observing their behaviours, made the ST aware of learning something.

Information from (behaviour of) others

By realising that I finally did what I planned to do

The moment that the ST had experienced that her strategy had been effective, she realised that she had learned something.

Experience of what works

Table 2.1 (Continued)

41

21

10

31

40

10

13

24

5

10

19

35

29

20

0.78

0.05

-0.38

0.64

0.56

0.23

-1.15

-0.91

-0.77

-1.37

-1.59

0.51

0.54

0.25

-0.02

-0.06

-0.10

-1.00

0.42

-0.51

0.59

0.27

-2.58

-0.26

-0.03

-1.26

0.41

0.85

It was a pity that I did not hear a lot of new things in this lecture Well, I did it, despite the fact that I had my doubts beforehand

The ST showed that someone else could have done better.

ST evaluated that she was satisfied about what she had learned.

No answer had been given.

Learning process under control of others

Evaluation of learning content

No answer

I want to grow more as a teacher and be more satisfied about myself

The ST formulated new learning goals or directions for her development.

The ST notified that she wants to maintain the same (teaching) behaviour.

The ST wanted to improve (more) what she had learned so far in the same direction.

The ST wanted to apply what she had learned to practice, possibly with some small adaptations.

The ST kept on doing the same as she did before, she had no specific plans

No answer had been given.

Formulating new goal/wish

Consolidation

Improving practice

Applying to practice

No specific changes

No answer

2

6

37

22

27

9

30

21

14

8

20

8

-

-0.05

-1.16

0.58

0.40

-0.07

0.70

0.22

-1.16

-1.94

-0.33

-0.07

-

0.11

-0.34

-0.31

0.64

0.45

-0.67

-0.71

-1.75

0.50

0.64

0.61

1

ST = Student teacher. Loadings in Bold are the one-third highest-scoring categories of a dimension and loadings underlined are the one-third lowest-scoring categories of a dimension. 2 ‘She’ and ‘her’ may also be replaced by ‘he’ and ‘his’. 3 The questions about learning goal orientation, notions of self-efficacy and argumentation for a strategy the total frequency could be skipped in case of an unplanned learning experience. These unintentional learning experiences were on these variables handled as missing value in the multiple correspondence analysis and consequently have no loadings on the two dimensions. 4 For two of the 50 unplanned learning experiences, this question turned out to be also relevant and was thus answered by the student teachers. Therefore, the frequency of this category is 48 instead of 50.

Just trying again!

I want to apply these ideas in a new group

I want to be clearer in my assignments by changing to written instructions

The next meeting I want to try again to discuss problems thoroughly

I will not interrupt them in the next lesson for unimportant things

The ST described in terms of concrete actions what she had been planning to do next.

Action plan

H Forethought on a new learning experience: ‘How will you proceed with this learning experience?’(Question 8)



I am very happy with it!

The ST only showed that there is nothing that she would have changed and that she was completely satisfied without giving more information why.

Completely satisfied

I should have followed this lecture in an earlier stage

ST evaluated the moment of this learning experience in her development.

Evaluation of moment of learning

Table 2.1 (Continued)

Chapter 2

39

The student teachers were asked to describe their learning strategies (variable C in Table 2.1). Five qualitatively different learning strategies could be discerned. These included learning by doing, learning by interacting, learning by processing information, and learning by applying theory to practice. As a last strategy, student teachers also reported reflection or evaluation as a method of learning in combination with one of these learning strategies. In addition to the question about their learning strategies, student teachers were also asked to report their reason for using this strategy (learning strategy control (variable D)). This seemed to be a hard question for student teachers, because in 60 cases, this question remained unanswered. Actually, in only twelve Learning Reports was an argument for a way of learning given. The categories of the monitoring variable (variable E) included the different ways student teachers realised they had learned something. By answering this question student teachers could give the indicators they used to monitor their learning results. The categories of this variable included the same type of categories as the learning strategies; for example, they also used reflection and information from others as an indicator for learning. Student teachers started the Learning Reports with a reflection on their object of learning (variable F). This means that even though student teachers had the same learning experience (e.g. a workshop about classroom management), they described their object of learning on different levels. Eight qualitatively different kinds of reflections on the learning outcome were identified. These reflections varied from a description of their learning outcome as a rule of thumb, to descriptions of ‘knowing that’, ‘knowing how’, and ‘knowing why’, which described their personal theories of practice. The student teachers were asked to evaluate their learning experiences (variable G). They referred to different aspects of the learning experiences, like the moment of learning, and their learning strategy and the largest category was an evaluation focused on their own behaviour (N = 40). In 20 learning experiences student teachers were completely satisfied. The variable forethought (variable H) dealt with the student teacher’s plans to proceed with their learning in a new learning experience. Six different categories were identified, of which four are behaviour-oriented in the sense that they are focused on changing or maintaining behaviour on the basis of this learning experience. In nine cases a new learning goal was set and in six cases no new plans were made.

40

Chapter 2

2.3.2 Relations between the categories of description and underlying dimensions In order to answer the second research question, which was aimed at determining the relations between the categories of the different variables and the underlying structure in the data, a multiple correspondence analysis (MCA) was carried out on the complete set of 133 Learning Reports, and all 50 categories of the eight variables. As described above, the questions about learning goal orientation, self-efficacy notions and learning strategy control were not relevant for unintentional learning experiences. These unintentional learning experiences were handled as missing values in these variables instead of being coded with a specific ‘missing’ category. In this way, we could prevent these ‘missing’ categories being near each other because of co-occurrence and in this way influencing the MCA solution too strongly. A common criticism of MCA is that outliers caused by low-frequency categories can have considerable effect on the solution. To discover the influence of possible outliers, we interpreted several solutions with and without outlying objects and categories. This procedure resulted in the removal of one Learning Report from the initial analysis. The analysis resulted in a two-dimensional solution. The dimensions 1 and 2 had an eigenvalue of respectively 3.61 and 2.71 and explained 45.1% and 33.9% of the variance. The scree test (Cattell, 1966) showed that a third dimension would not add a substantial amount of extra explained variance to the solution; in addition, the interpretation of the third dimension was very comparable to the second dimension and did not provide extra meaningful information. As explained in the Method section, categories close together in the representation are associated with the same Learning Reports and are therefore related. In Table 2.1, the loadings of all categories on the dimensions can also be found. The one-third highest-scoring categories of each dimension are presented in bold type and the one-third lowest-scoring categories of each dimension are underlined. By describing the similarities of the regulation activities on both ends of the dimension, we can interpret the dimensions and thus the underlying structure in the data. The regulation activities scoring low on Dimension 1 are categories that show the presence of an external source of regulation in the learning process or reflect lack of regulation. Also, almost all ‘no answer’ categories of the different variables score low on the first dimension. The other low-scoring activities have in common that the learning is under the control of others (especially the 41

choice for the learning strategy and the evaluation) or under the control of an external source of information (monitoring of the learning results). The decisions in the learning process made by the student teacher him/herself are not well thought-out (superficial reflection and self-efficacy). On the other hand, the high-scoring categories show a more active pattern of regulation where the student teacher reflects more deeply on the learning content, the learning process and his or her own role in this. In these learning experiences, student teachers make their own decisions for a learning strategy, but also actively use information from others and reflect on that. Therefore this dimension represents passive regulation (a combination of external and no regulation) at one end of the spectrum and more active regulation of learning at the other. On the second dimension, the low-scoring categories are not well spread over all variables. Table 2.1 shows that the variables of the first phase of the learning process, namely goal orientation, self-efficacy beliefs and learning strategy control, have almost no low scoring categories. In the other variables, the low-scoring categories also show characteristics of deep processing (when combined with a high score (active regulation) on the first dimension) as well as some ‘no answer’ categories (when combined with a low score (passive regulation) on the first dimension). Examples of these deep processing activities are learning and monitoring by reflecting, and describing the learning object as ‘knowing about myself’ and ‘knowing why’. Furthermore, the category evaluation of the learning strategy is also positioned low on this dimension. The high-scoring categories on this dimension show a thorough orientation on the learning process with explicit goal-setting, different kinds of self-efficacy beliefs, and explicit arguments for the learning strategy. Highscoring categories of the variables of the reflection and evaluation phase of the learning process are, however, more superficial. The object of what has been learned is described as ‘knowing how’ or as ‘a specific teaching practice’, monitoring is based on what works, and only the moment of learning is evaluated or the student teacher is completely satisfied. Actually, at both ends of this dimension, we see important regulation activities for steering one’s own learning process. The high-scoring categories show regulation of learning in terms of planning and preparing learning opportunities. The low-scoring categories are characterised by unplanned learning experiences, but when combined with the active side of the first dimension these regulation activities show deep aspects such as reflection and evaluation of the learning process in 42

Chapter 2

order to learn from these unplanned experiences. Therefore, we interpreted this dimension as differentiating prospective from more retrospective regulation of learning.

2.3.3 Differences between two learning contexts The reported learning experiences occurred in a lot of different situations: for example, during lectures, teaching, informal talks with colleagues or peers, mentoring conversations, or simply at home. To answer the third research question, concerning the differences between the two different contexts (university and practice schools) of the dual learning programme as regards the nature of SRL, all learning experiences were coded according to the learning context in which they took place as described by the student teachers in the Learning Report. Three different contexts were distinguished: learning at the practice school (N = 82), at the teacher education institute (N = 45), and in contexts outside these two places, for example at home or in a library (N = 6). The average positions of the learning experiences in the different contexts on the two dimensions have been calculated and are shown in Table 2.2. Table 2.2 Means (M) and Standard Deviations (SD) on Dimension 1 (activeness) and Dimension 2 (prospectiveness) for different contexts of the learning experience Context of the learning experience

Scores on Dimension 1 M (SD)

Scores on Dimension 2 M (SD)

Teacher education institute

-0.96 (1.11)

-0.21 (1.17)

Practice school

0.48 (0.70)

-0.16 (1.26)

Other context

0.42 (0.76)

0.19 (0.29)

These results show that the learning experiences at the practice school and in other contexts score much higher on Dimension 1 than learning at the teacher education institute (F(1,130) = 41.03, p < .001). From this we can conclude that passive and active regulation are related to the context in which the learning occurs: learning experiences from the teacher education institute context show more passive regulation activities, whereas learning experiences from the practice school and other contexts show more active regulation activities. On Dimension 2 no differences are apparent (F(1,130) = 0.29, p = .75), indicating no

43

differences between the learning contexts in terms of prospectiveness or retrospectiveness of regulation.

2.4 Conclusion and discussion The aim of this study was to describe the nature of self-regulation of learning in a dual learning programme such as teacher education. Three questions were posed concerning the variety and nature of student teachers’ regulation activities, the relation among these regulation activities in underlying dimensions, and the differences in regulation activities between the different contexts of a dual learning programme. Content analysis of 133 Learning Reports resulted in eight variables and five to seven categories per variable. Since the framework of Pintrich (2000) was used to develop the instrument, the main elements of SRL described in his model were also used as variables for the content analysis. The large number of categories, however, reflects the large variation in regulation activities that student teachers showed across different learning experiences in a dual learning programme. Emerging results from the descriptive categories were extensive focuses of the regulation activities on teaching practice and changes in own behaviour and a high frequency of spontaneous and unplanned learning experiences. Most models of SRL are based on the assumption that learners consciously direct their activities towards a learning goal (Boekaerts & Corno, 2005; Pintrich, 2000; P.H. Winne, 1995). The results from the present study resemble earlier observations that in informal learning the goals tend to be broader, which may result in more variability in what gets learned (Boekaerts & Minnaert, 1999). Teachers’ goals are more focused on the achievement and well-being of their students than on their personal achievement (Butler, 2007). Although student teachers focused their regulation activities more on their behaviour, for example on attaining working goals, choosing strategies for their pupils’ learning instead of their own learning, or monitoring in terms of ‘what works in the classroom’, these activities all resulted in learning for themselves. Further studies have to show whether working goals are as beneficial for student teacher learning as setting personal learning goals. The solution of the MCA showed the underlying structure in two dimensions in these data. These dimensions were interpreted as passive versus active regulation and prospective versus retrospective regulation. The combination of these two dimensions results in four different kinds of 44

Chapter 2

regulation: active prospective regulation, whereby the student teacher actively searches for, and plans learning experiences; passive prospective regulation, whereby the learning experiences are planned and the learning processes are set up mostly by others; active retrospective regulation, whereby student teachers learn from spontaneous experiences by actively monitoring, reflecting on and evaluating them; and passive retrospective regulation in which the learning experiences are not planned by the student teacher and during and after the learning experience no or only a very superficial method of regulation can be seen. Although in previous research the dimension self-regulation versus external regulation is often used to describe variation in regulation of learning (Kaplan, 2008; Vermunt, 1998), this distinction turned out to be less relevant for the present data. First, we found on the low side of the first dimension not only external regulation, but also lack of regulation. Furthermore, on the active side of the first dimension categories were found that showed the influence of others on the learning process. For example, for the monitoring of their learning results, active regulating student teachers often needed the information of others. Although the labels of the dimension self versus external regulation suggest that the ‘external’ can be any other person except the learner him- or herself, in previous studies into student learning this ‘external’ often referred to a teacher (Vermunt & Verloop, 2000). Researchers have hardly ever taken into account other external factors regulating learning like peers or other contextual factors, except for some studies into regulation of learning in collaborative learning settings (Volet, Summers, & Thurman, 2009). For student teachers, in many learning experiences no teacher educator is present who can externally regulate their learning. Nevertheless, there are still many others present during the learning experiences like colleagues, peers or pupils who can consciously or unconsciously steer the student teachers’ learning. By naming this dimension passive versus active regulation, we can describe all variations in the activity of the student teachers in regulating their own learning, despite or owing to the presence of others. The second dimension showed how learning experiences varied in the focus of regulation. Although regulation of learning embraces all phases of the learning process, this dimension showed that in the reported learning experiences the focus was either more on prospective regulation, in other words, the planning and goal-setting phase, or on retrospective regulation involving the monitoring, reflection and evaluation phase. According to 45

Boekaerts and Corno (2005), for the most part researchers narrow the scope of students’ capability to regulate their learning through a focus on the academic side of education: namely learning and achievement goals. By describing the regulation strategies that learners use to reach academic goals, SRL theories reveal little about students’ actions and efforts to regulate when they are not so mindfully engaged in learning (Boekaerts & Corno, 2005). Although we know that setting learning goals is related to more effective academic learning (Schunk & Zimmerman, 1998), we also know that implicit learning is a large part of the learning process of becoming a teacher. Although teacher learning can also be planned, much of their learning is informal in the sense that it is ongoing and often incidental (Hodkinson & Hodkinson, 2005). The findings of the study of Van Eekelen and colleagues (2005) showed in their study that when learning experiences are classified as self-regulated only when they have pre-set learning goals, less than one-third of experienced teachers’ learning experiences could be classified as being self-regulated in nature. In the field of workplace learning it is well known that people learn intentionally as well as unintentionally (Tynjälä, 2008). Eraut (2004) makes a distinction between deliberative, reactive and implicit learning. The distinction between deliberative and reactive learning is particularly comparable with the differences in regulation found within the second dimension. The interpretation of the second dimension showed that learning experiences starting as reactive and non-deliberate, lacking a learning goal, can still involve deliberate, active regulation activities, though in a more retrospective way. Therefore, we argue that for this context setting goals is not a prerequisite for active regulation of learning. Moreover, the fact that this dimension separates active prospective and active retrospective aspects of regulation implies that in this context active prospective and active retrospective aspects do not occur often together in the same learning experience, showing that regulation of student teachers’ learning processes is less sequenced than SRL theories prescribe (Azevedo, 2009). The first dimension, representing the activeness of regulation, turned out to be related to the context of the practice school. Learning at the workplace is often less deliberate in nature (Tynjälä, 2008) and also the context of the practice school is often not primarily structured to stimulate the learning of the (student) teacher, but that of the pupils instead. From this perspective we would expect more reactive learning and more passive regulation during learning at the practice school and more active regulation when there is time 46

Chapter 2

for learning as it is at the teacher education institute. An explanation of the opposite finding in this study could be that at the practice school there is no external regulator for student teachers’ learning process, so everything that is learned has to be self-initiated in some way. In this study, student teachers were asked to report several self-chosen learning experiences from different learning contexts. The advantage of this design is that student teachers could report learning experiences very close to the moment that they occurred and, if necessary, after some time for reflection. Also, the openness of the instrument made it possible to fill in every kind of learning experience, and this resulted in a very broad overview of student teachers’ regulation of learning in various learning experiences in a dual learning programme. The instrument also has some limitations; for example, it may have underrepresented the regulation of the process of development spread over multiple experiences. Although student teachers were instructed to describe learning experiences on different levels, the focus on the learning experience may have caused reports of learning experiences limited to one experience. Furthermore, to use the instrument it was necessary for student teachers to be aware of something they had learned. Therefore, it is possible that experiences, from which the learning results evolved only after some time were reported less often. The study has been conducted at only one teacher education institute; further research is necessary to confirm the existence of the different dimensions in a larger population. The set of categories identified in this study shows the variation in student teachers’ regulation of learning. The dimensions underlying these categories gave better understanding of the concept of self-regulation of learning in the context of dual learning programmes such as teacher education. This contributes to the understanding of the phenomenon of self-regulation in the context where learners need it most, in the absence of a structured learning environment. These results make it possible for future studies and practice to describe and improve the quality of student teachers’ regulation of learning in a more nuanced and relevant way for this specific context.

47

dimensions results

examples overview

institute

meaning-oriented focus doing

active

passive

complex situations

orientations

theoretical

become

reflection

significant

below

weekly

described collected

perspective

insight received

goals

definitions

vermunt

metacognition

previous

getting

trying

turned

consisted

male instructional understanding

responsibility

focused experienced correspondence reproduction-oriented meaning detailedparticipants asked

goal-setting

existing

months yourself

classified

planned

stimulated measurement spontaneous

identify

across

approach

positioned university

leeuw

applied pre-service

things directed

dual

requires

female

frequencies

strategy

learner

experiences

education

prospective

preferences

terms

reports

evaluation

characterised

unplanned

teaching involves explanation

phase

self-regulated

regulation attention

performance types fourteen

maria

individual

following

development

coherent

something

research

observing

monitoring

knowledge

important

typical

separates

peers

defined

obtain

choice

average

secondary

clearly

quitted

student teachers' retrospective

interviews

vermetten

its

included

data

beginner

academic

constructs

increasing

component

srl

kinds

interpretative

chapter

aspects

programme

dissonant

questions

qualitative

improving

personal

sample

oosterheert

framework

limited

particularly

almost

reference

deliberate

concerning

discussion

learning

versus

conceptions

coded

cognition

structure

environments

theory

students

stable

multiple

quadrants

instruments

relation

task

self-efficacy

information lessons

variables

teacher works

explained

categories

background

problems

categorised learning reports analysis remained description context

different

looking

suggestions really

outcomes

combination

dynamic

various

variation

lack

3 Student teachers’ self-regulated learning: An examination of regulation patterns and conceptions of learning in a dual learning programme3 Abstract University-based teacher education is nowadays often organised in dual learning programmes in which studying at university and learning from practice in schools are combined. This study was aimed at increasing our understanding of student teachers’ self-regulated learning in such programmes. Cross-sectional data have been collected from 28 student teachers about their learning conceptions and regulation activities through interviews and six open question logs on self-chosen learning experiences. Most student teachers showed multiple types of regulation while learning in a dual learning programme, and almost all student teachers showed at least once within their six learning experiences an active way of regulating their learning. The student teachers were categorised as having a meaning-oriented or a reproduction-oriented learning conception. The relation between student teachers’ learning conception and regulation pattern turned out to be more complex than has been suggested by academic learning theories.

This chapter has been submitted in adapted form as: Endedijk, M., Brekelmans, M., Verloop, N., & Vermunt, J.D. (under revision). Student teachers’ self-regulated learning: An examination of regulation patterns and conceptions of learning in a dual learning programme. 3

49

3.1 Introduction University-based teacher education is being organised more and more in dual learning programmes in which two types of learning environments are combined: studying at university and learning in practice. These programmes often require a high degree of self-regulated learning (SRL) of student teachers: they have to self-evaluate their competences, clarify their learning needs, formulate personal development plans, document their learning progress in a portfolio, reflect on their learning, adjust their learning processes, and combine knowledge gained from teaching experiences and university courses. Moreover, from the perspective of lifelong learning (Cornford, 2002) and because teachers as experts in learning have to act as role models for their students, it is very important that student teachers learn to regulate their own professional learning (Hagger, Burn, Mutton, & Brindley, 2008; Kramarski & Michalsky, 2009; Tillema, 2000). Setting up a teacher education programme that requires SRL, however, does not automatically lead to the development of excellent self-regulated learners (Evensen, Salisbury-Glennon, & Glenn, 2001). Students starting a postgraduate teacher education programme in The Netherlands already have a master’s degree in a specific subject area and thus have demonstrated ample proficiency in academic learning. Nevertheless, learning in a dual learning programme requires different learning skills: student teachers learn from many different sources of information, and daily practice at the workplace has a big influence on their learning processes. Learning at the workplace is less intentional and planned, and more contextual and collaborative, than learning in educational learning settings (Tynjälä, 2008). Furthermore, teacher learning does often not have preset objectives or easily identifiable outcomes (Hodkinson & Hodkinson, 2005). Also, previous studies have shown that changes in learning environments also often lead to temporary friction or dissonance in students’ learning approaches (LindblomYlänne & Lonka, 2000). Thus, it is not clear yet how student teachers manage to regulate their learning in these new and demanding dual learning programmes. Therefore, this article aims at increasing our understanding of student teachers’ self-regulated learning in a dual teacher education programme.

50

Chapter 3

3.2 Theoretical framework Recently, Dinsmore, Alexander, and Loughlin (2008) gave a historical review of how metacognition, self-regulation, and self-regulated learning have been defined. One of the conclusions of their review was that the conceptions of these constructs are not fixed but that they continue to move and the concepts have become increasingly intertwined. The term metacognition is attributed to Flavell (1976) who started the work on metacognition by introducing the term metamemory (Flavell, 1971). He defined metacognition as “one’s knowledge concerning one’s own cognitive processes and products or anything related to them, e.g. the learning-relevant properties of information or data” (Flavell, 1976, p. 232). A second tradition in research on metacognition is based on the work of Brown (1978), who used the term metacognition for self-regulatory mechanisms including such skills as predicting, checking, monitoring, reality testing, and coordination and control of deliberate attempts to study, learn, or solve problems. The concept of self-regulated learning (SRL) was advanced by Corno and Mandinach (1983) in order to reach a more integrated theory of learning. The review by Dinsmore, Alexander, and Loughlin (2008) shows that two definitions of self-regulated learning (SRL) research prevail today. There are studies which use the term SRL, but are actually engaged in research on selfregulation (i.e. the dynamic self-regulatory skills without the stable knowledge part) of learning processes. The second definition of SRL research is applied to research which is intended “to bring elements of metacognition and selfregulation together, that is, to span both research traditions” (Dinsmore et al., 2008, p. 405). Therefore, whereas metacognition was originally used to describe knowledge about cognition (Flavell, 1976) as well as self-regulation of cognition (Brown & DeLoache, 1978), nowadays the term SRL is used to describe the integration of students’ more stable knowledge about learning with the more dynamic part described as self-regulation. In line with the second definition of Dinsmore et al. (2008), the present study will focus on both parts of SRL: the more stable part of SRL, conceptualised as learning conceptions, and the more dynamic self-regulation activities of student teachers. Below, both parts of SRL and the relation between them will be described in more detail.

3.2.1 The dynamic regulation component of SRL: self-regulation activities In most theories, self-regulation of learning is defined as an “active, constructive process whereby learners set goals for their learning and attempt 51

to monitor, regulate and control their cognition, motivation, and behaviour, guided and constrained by their goals and contextual features in the environment” (Pintrich, 2000, p. 453). This process consists of different phases that represent a general time-ordered sequence that individuals are likely to go through as they perform a task, although different phases can occur simultaneously. Phase 1 involves planning and goal-setting as well as activation of perceptions and knowledge of the task and context and the self in relation to the task. Phase 2 concerns various monitoring processes that represent metacognitive awareness of different aspects of the self and task or context. Phase 3 concerns efforts to control and regulate different aspects of the self or task and context. Finally, Phase 4 represents various kinds of reactions and reflections on the self and the task or context (Pintrich, 2000). Research has demonstrated that both teachers and students draw on SRL strategies to cope with situational demands in the classroom (Randi, 2004). Research into SRL has, however, focused primarily on student learning in traditional instructional settings. Most studies on student teachers or experienced teachers deal with the issues of how teachers can learn how to promote active and self-regulated learning in their students (Bolhuis & Voeten, 2001; Kramarski & Michalsky, 2009; Niemi, 2002; Perry, 2008; Perry, Phillips, & Dowler, 2004) and are not focused on teachers’ regulation of their own learning. A few studies have described regulation processes of student teachers in a pre-service teacher education programme (e.g., Corrigan & Taylor, 2004; Järvenoja & Järvelä, 2007), but these student teachers are no different from regular college students in the sense that they just follow a course at the university and are not involved in teaching in practice at the moment of investigation. No empirical studies were found describing student teachers’ regulation activities while learning from study at university and learning from practice. Studies into experienced teachers’ self-regulation, however, have shown that academic staff vary in their degree of engagement in self-regulation in learning; differences were found particularly in the extent to which these teachers set explicit goals for themselves (Kreber, Castleden, Erfani, & Wright, 2005; Van Eekelen, Boshuizen, & Vermunt, 2005). In the study by Van Eekelen and colleagues (2005), less than one-third of experienced teachers’ learning experiences could be categorised as self-regulated learning in that they were planned beforehand. In order to gain more insight into the nature of student teachers’ selfregulation, we conducted a study on the variation in student teachers’ 52

Chapter 3

regulation activities in different learning experiences in a dual learning programme (see Chapter 2). This study has shown a large variation in student teachers’ regulation activities which can be described in terms of an underlying structure of two dimensions: the first dimension discerns passive regulation from active regulation of learning and the second dimension separates prospective and retrospective regulation of learning. With the help of these dimensions every learning experience could be described in terms of its activeness or passiveness and whether the focus of the regulation is more prospective or more retrospective. The present study will use these dimensions to describe the variety in student teachers' regulation activities. For a detailed description of the various regulation activities and the interpretation of the two dimensions, we refer the reader to Chapter 2.

3.2.2 The stable component of SRL: Conceptions of learning Student learning has historically been studied from two perspectives, the student approaches to learning (SAL) perspective and the information processing (IP) perspective. Although research on metacognition has its origin in the IP perspective (later also called the self-regulated learning perspective), research on the knowledge people have about learning is certainly not limited to this perspective. The SAL perspective has also studied students’ knowledge about learning in terms of their learning conceptions, which included students’ conceptions about learning in general, their learning and thinking activities, oneself as a learner, and their regulation conceptions (Vermunt, 1996). The SAL perspective started with the work of Säljö (1979), who used phenomenographic methodology to describe students' conceptions of learning. He found a hierarchy of five learning conceptions in which learning was conceptualised by students as: (a) increasing one’s knowledge, (b) memorising and reproducing, (c) applying, (d) understanding, (e) seeing something in a different way. This taxonomy was confirmed by Marton, Dall'Alba, and Beaty (1993) and a sixth conception was added: (f) changing as a person. The first three describe quantitative dimensions of learning in which learning is seen as reproduction of material and reflects a surface approach, and the latter three are more qualitative and focus on the role of meaning in learning, reflecting a deep approach (Marton et al., 1993; Säljö, 1979). This central distinction between learning as reproduction with a surface approach and meaning-oriented learning reflecting a deep approach has remained a core construct for decades

53

in studies into students’ learning in different learning contexts (Lonka, Olkinuora, & Mäkinen, 2004). In the context of pre-service teacher education, Oosterheert and Vermunt (2001) used the concept of learning orientation to denote typical combinations of learning conceptions and preferences for learning- and regulation activities. They found five different orientations towards learning to teach. The main dimension in which these orientations differed was also meaning-oriented versus reproduction-oriented learning (Oosterheert & Vermunt, 2001). The authors define this typology as follows: Reproduction-oriented student teachers are directed to the improvement of performance through the gathering of “cut-and-dried” practical suggestions; “How to...?” and “What works?” reasoning is dominant in their learning. These student teachers do not question the limitations of their existing interpretative frame of reference and (therefore) do not experience non-understanding. The suggestions they gather remain unconnected and of equal importance. Meaning-oriented student teachers are aware that their interpretative frame of reference is limited. They are also directed at improving their performance through the acquisition of a better understanding of teaching and learning. Apart from procedural information, they are also interested in conceptual information that clarifies and structures their experiences (2001, p. 149).

3.2.3 Relations between learning conceptions and self-regulation activities Conceptions of learning are considered to influence how students approach learning in particular contexts (Brownlee, Purdie, & Boulton-Lewis, 2001; Meyer, 1999). This is especially true of ill-defined problems and complex tasks (Pieschl, Bromme, & Porsch, 2008), as is often the case in a dual learning programme such as teacher education. In the field of student learning, studies have shown that a meaning-oriented way of learning leads to higher learning outcomes than a surface-oriented way of learning (Vermunt, 2005). In the last decade, more attention has been given to incoherent patterns of learning (e.g., Meyer, 2000), since people often show inconsistency in their learning conceptions and learning activities (Lindblom-Ylänne, 2003; Van Petegem, Donche, & Vanhoof, 2005; Vermunt & Minnaert, 2003).

54

Chapter 3

Patterns of student learning can be typified as dissonant when students’ learning conceptions are not congruent with the learning activities or when students show aspects of multiple contradictory learning approaches at the same time (Cano, 2005; Vermunt & Verloop, 2000). This is, for example, the case when students do not show the combination of self-regulation and a meaning-directed learning approach, or the combination of external regulation and a reproduction-oriented learning approach, which are the combinations that occur most often and are theoretically coherent (Beishuizen, Stoutjesdijk, & Van Putten, 1994; Vermunt, 1998). A dissonant learning pattern is often related to poor academic performance (Beishuizen et al., 1994; Cano, 2005; Vermunt & Verloop, 2000). Lindblom-Ylänne (2003) showed, however, that whereas clearly dissonant learning patterns indeed were related to study problems, slightly dissonant learning patterns could be seen as good study practices that had been changed by the environment. Particularly when students enter a new learning environment, there may be a temporary misfit between their learning conceptions and the learning activities that are required in the new learning environment (Brownlee, Purdie, & Boulton-Lewis, 2003; Vermunt & Vermetten, 2004). We expect that a dual learning programme in itself, where student teachers are continually changing between learning environments, may also evoke dissonance. These dissonances may appear between learning conceptions and regulation activities, but student teachers may also show a dissonant variation in their regulation activities across different learning experiences.

3.3 The present study Because of the possibility of dissonance between the two parts of SRL, this study examines both student teachers’ regulation activities and their learning conceptions at the beginning and end of a teacher education programme. A small-scale qualitative approach is adopted. The limited number of participants allowed us to study student teachers’ regulation activities in multiple learning experiences in the different contexts of the dual learning programme. More specifically, we aimed to answer the following research questions: 1. What are patterns of student teachers’ regulation activities across different learning experiences in a dual learning programme? 2. What are student teachers’ conceptions of learning to teach? 55

3. How are student teachers’ regulation patterns and conceptions of learning related? 4. What are the differences between beginner and experienced student teachers in terms of their regulation patterns and learning conceptions? This study was conducted in a postgraduate pre-service teacher education programme; students first obtain their master’s degree in a specific subject area and then enter the one-year programme to obtain their teaching degree for secondary education. During the programme, student teachers attend weekly classes at the university, while also doing teaching practice at schools or holding a paid job as a teacher. Because of the shortage of teachers in The Netherlands, a lot of student teachers already have a job or have applied for a job when starting the teacher education programme. Although they are still unqualified, they start from the first day as a teacher at a secondary school. The other student teachers are more gradually exposed to the teaching profession. They start by observing other teachers and peers and taking over some lessons from an experienced teacher and only after six months do they become responsible for all aspects of teaching. All student teachers have much freedom in designing their personal curriculum, based on their prior experiences and concerns. They keep a portfolio in which they make self-evaluations and a personal development plan to direct their learning processes. In this personal development plan they also reflect on their own responsibility for their learning processes.

3.4. Method 3.4.1 Participants Twenty-eight student teachers from one Dutch postgraduate academic teacher education programme volunteered to participate in this study. The student teachers were selected in a stratified sample intended to capture as much variation as possible in the following variables: teaching experience, supervising teacher educator, and school subject. In order to see differences between students at different stages in the programme, fourteen beginners (two to four months after the start of the programme) and fourteen experienced student teachers (six to nine months after the start of the programme) were asked to volunteer. Furthermore, both student teachers with a paid job and student teachers doing an internship were 56

Chapter 3

included in the study. Seven student teachers who were not willing to participate (mostly because of lack of time) were replaced by other student teachers with the same characteristics in terms of these variables. The final set of participants consisted of nine male and nineteen female student teachers from all school subjects. Five student teachers already had prior experience of teaching at a secondary school, four of them for fewer than six months and one of them for ten years at the lower level of secondary school. At the different stages of the programme, the number of lessons student teachers had to give on their own varied. The beginners' group was teaching an average of 4.4 lessons per week (SD = 2.8), whereas experienced student teachers taught an average of 10.1 lessons a week (SD = 2.8).

3.4.2 Instrumentation For the measurement of student teachers’ regulation activities an open question log called Learning Report was developed. Student teachers were asked to describe six self-chosen learning experiences, stimulated by ten open questions about different regulation activities. The questions of the Learning Report were developed on the basis of the conceptual model presented by Pintrich (2000), which represents a blueprint for the development of new instruments for measuring SRL. For every phase of SRL, as described in the theoretical framework, two or three questions were included in the Learning Report. For the planning and goal-setting phase, questions were inserted about student teachers’ goal orientation, notions of self-efficacy, learning strategy and forethought. The questions concerning the monitoring and control phase described student teachers’ control of the learning strategy and monitoring of the learning results. The questions concerning the reflection phase were focused on student teachers’ reflection on the learning outcome and evaluation of the learning experience (see Chapter 2). For the measurement of student teachers’ learning conceptions, a semistructured interview was developed. Questions regarding student teachers’ learning conceptions were selected from the interview schedule as develop by Oosterheert and Vermunt (2001). The central question for our interviews was: How do you learn to become a teacher, and can you describe this process?

3.4.3 Procedure The fourteen experienced student teachers, who had started the teacher education programme at the end of August, were interviewed between mid57

February and mid-March 2006 (six or seven months after the start of the programme) and their Learning Reports were collected between the end of February and the beginning of May 2006 (six to nine months after the start). The fourteen beginner student teachers had started the teacher education programme in February. They were interviewed in April (two or three months after the start) and the Learning Reports were collected between the end of April and the end of June (two to four months after the start). After the interview, all student teachers received the instructions for using the Learning Reports. Student teachers were asked to use a learning experience no more than a week old for the Learning Report; this could be any kind of experience that was part of their development as teachers. Because of the diversity of student teachers’ learning processes, as described previously, the student teachers were stimulated to describe different kinds of learning experiences: learning experiences that had taken place at the teacher education institute as well as at the practice school, planned as well as spontaneous learning experiences, and both successful and unsuccessful learning experiences. It was explained that in the case of an unplanned learning experience, some questions were not relevant. The student teachers were therefore instructed to answer these questions with not relevant. To stimulate student teachers to complete and send back the reports weekly, they received a weekly reminder by e-mail, and they were phoned when they had not responded within two weeks. In total, 134 Learning Reports were collected, eighteen student teachers completed all six Learning Reports as required, and four student teachers quitted after four or five Learning Reports. Two student teachers did not send in more than three Learning Reports, and three others did not complete a single report. One student teacher’s Learning Report was excluded from analysis, because the reported experience was not related to learning to teach (N = 133). The reasons student teachers gave for failing to complete all six reports were mostly illness, quitting teacher training, or lack of time.

3.4.4 Data analysis In our previous study, the Learning Reports were analysed in two phases. The first step of the analysis was a phenomenographically informed content analysis performed in order to reach a stable set of categories covering the qualitatively different descriptions of self-regulated learning reported in the Learning Reports in as few categories as possible. During this process eight 58

Chapter 3

variables were established, each representing a different aspect of the regulation process and one of the questions of the Learning Report, around which a set of five to seven categories per variable could be constructed. All text fragments of the Learning Report were coded in terms of this set of categories. From the total of 1197 fragments that were coded, 10% were also coded by an independent second researcher to compute inter-rater reliability. After a small adaptation of the code book, this resulted in an overall Cohen’s Kappa of 0.90. In the second phase, multiple correspondence analysis (MCA) was used, also referred to as homogeneity analysis (Mair & De Leeuw, 2008), which is a form of nonlinear principal component analysis (De Leeuw, 2005). This technique makes it possible to discover relationships between categorical data and the underlying structure in the data, and is particularly useful when there are many variables and categories (Mair & De Leeuw, 2008). The outcome of the MCA shows the position of the categories on dimensions, which are used for the interpretation of these dimensions. The first dimension was interpreted as passive regulation versus active regulation of learning and the second dimension was interpreted as prospective versus retrospective regulation of learning (see also Figure 3.1). Both the categories and the objects, in this case the Learning Reports, are also positioned on the dimensions. This resulted in six scores on the two dimensions for every student teacher or fewer if they had not submitted all the Learning Reports. For all student teachers, graphs were made of their six Learning Reports positioned on the two dimensions. Only those who reported four or more learning experiences were included in the analysis (N = 22). As can be seen in Figure 3.1, four quadrants were made to categorise the learning experiences into one type of regulation. The quadrants of Figure 3.1 are separated by the axes of the dimensions through the origin, since multiple correspondence analysis positions all objects (learning experiences) on both dimensions in such a way that the average on the dimensions is close to zero.

59

PROSPECTIVE REGULATION

2

4

3

ACTIVE REGULATION

PASSIVE REGULATION

1

RETROSPECTIVE REGULATION

Figure 3.1. Dimensions of student teachers’ regulation of learning. The data of the interviews were analysed in order to describe the student teachers’ different learning conceptions. The categorisation of the interviews was based on the typology of meaning-oriented and reproductionoriented learning of student teachers as described previously in the theoretical framework (Oosterheert & Vermunt, 2001). Keywords that were used to identify meaning-oriented learning conceptions were: learning by deepening understanding, gaining insights, reflecting, looking for structure, looking for information from others, looking for theoretical information, and looking for food for thought. Keywords that were used to identify reproduction-oriented conceptions were: learning by doing, trying again, keeping what works, looking for tips and tricks. All interviews were coded according to these definitions by two researchers. The first independent categorisations led to an agreement of 79%. The cases that were not agreed upon turned out to be cases in which student teachers gave examples of both conceptions. In order to decide what their dominant learning conception was, we took into account the following guidelines: all student teachers learn by doing, but for meaningoriented student teachers this is not their only way of learning; reproductionoriented student teachers also use the word reflection but in terms of remembering experiences and writing these down in their portfolio instead of critical reflection activities; reproduction-oriented student teachers also 60

Chapter 3

mention that a more meaning-oriented learning conception is preferred by their supervisors, but they still do not see it as their conception of learning. Discussion between the two researchers and the involvement of a third expert on student teacher learning led to a final agreement on all cases.

3.5. Results 3.5.1 Patterns of student teachers’ regulation activities across different learning experiences The student teachers showed considerable variation in their regulation of learning across their six learning experiences. For all student teachers, the six learning experiences were spread over multiple quadrants. In order to explain the meaning of these quadrants more precisely, we will first give a more detailed description of the interpretation of the two dimensions and show some examples from the data. A learning experience that was passively regulated by student teachers was characterised by lack of argumentation for decisions they had made as well as answers that showed that someone else was in charge of the learning process. Furthermore, many aspects of the regulation process were not described at all. An actively regulated learning experience showed a more active pattern of regulation where the student teacher reflected more deeply on the learning content, the learning process and his or her own role in it. The student teachers made their own decisions for a learning strategy, but also actively used information from others and used it in their reflections. In a prospectively regulated learning experience, the focus of the regulation activities was on the first phase of the learning process. The learning experience has been planned, goals have been set and an argumentation for choosing a learning strategy has been given. The phase after the learning experience received less attention; the monitoring, reflection, and evaluation were more superficial. The retrospectively regulated learning experiences were often unplanned, so no goal-setting or deliberate thinking about learning strategy and self-efficacy had taken place. The regulation focused on the monitoring, evaluation and reflection part of the learning process. Below, the four quadrants are described, including their frequencies in the data. Examples of learning experiences of the four quadrants are shown in Figure 3.2.

61

1 Passive prospective regulation (N = 25, 18.8% of the total set of 133 learning experiences): this learning experience was planned, but the choices in the learning process were often made by others and not well thought out by the student teacher. The focus in the regulation was on the starting phase of the learning experience. The example in Figure 3.2 is that of an experienced female student teacher of the German language. 2 Active prospective regulation (N = 47, 35.3%): this learning experience was actively set up by the student teacher, but the focus remained on the start-up phase of the learning experience, and the reflection phase of the learning process received less attention. The example in Figure 3.2 is that of an experienced male student teacher of scriptural knowledge. 3 Active retrospective regulation (N = 37, 27.8%) : this learning experience was often unplanned, or not well thought out in advance by the student teacher, but the active regulation took place during evaluation and reflection on the learning experience. The example in Figure 3.2 is that of an experienced female student teacher of music. 4 Passive retrospective regulation (N = 24, 18.0%): this learning experience could be characterised by its lack of regulation, as it was often unplanned, and student teachers only realised afterwards that they had learned something, but without any active evaluation or reflection. The example in Figure 3.2 is that of an experienced male student teacher of biology.

62

Chapter 3

PASSIVE REGULATION

What did you learn? That students are also busy with other things than just school. Did you plan to learn this, and if so, why did you want to learn this? No, I realised it during a meeting with colleagues. There are a lot of students who have other things on their mind which explains their attitude at school. Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this? Not relevant. How did you learn this? [No answer] Why did you learn it in this way? [No answer] How did you realise that you had learned something? I realised it after the meeting. If you look back, about which aspects are you satisfied and what would you do differently next time? It makes you aware of other things. How will you proceed with this learning experience? I will write this down in my portfolio and try to realise more often what else students’ have on their mind.

ACTIVE REGULATION

What did you learn? I learned how to did you learn? I tried to change PROSPECTIVE REGULATION What the climate in the classroom from deal with my pupils’ speaking fluency chaotic towards a more structured working condition. in the classroom. I had not much idea about how to teach speaking fluency (…). Did you plan to learn this, and if so, why did you want to Did you plan to learn this, and if so, why did you want learn this? I planned to do this, because I thought that there was a lot to improve in the working climate of this to learn this? I planned to learn at least something during classroom. this lecture. My students are not motivated for speaking fluency, they don’t care. Maybe it has to do with my Did you expect to succeed in learning this and what instruction. made you think you would (not) succeed in learning this? I was not quite sure, but I wanted to do everything Did you expect to succeed in learning this and what possible to make this work, besides, I had success before made you think you would (not) succeed in learning this? I had the feeling I was going to get some good ideas, with these students. How did you learn this? Just by doing what I prepared to because that is the goal of this kind of lecture and in the do; communicating in a very clear way to this class how I past I always have learned something during lectures. How did you learn this? The teacher educator used a wanted them to behave differently from my previous strategy during the lecture that we can also use ourselves lesson. Why did you learn it in this way? [No answer] in the classroom. How did you realise that you had learned something? I Why did you learn it in this way? The teacher educator had chosen this strategy. realised I had learned something, since my pupils indeed were more quiet and cooperative. How did you realise that you had learned something? If you look back, about which aspects are you satisfied We worked with new materials. Everything was new. If you look back, about which aspects are you satisfied and what would you do differently next time? In general, I am satisfied with this learning experience; however, I had and what would you do differently next time? It is good one weak moment in which I didn’t act according to to get new ideas, I would not have changed the school rules. anything How will you proceed with this learning How will you proceed with this learning experience? In the future, I want to continue teaching in experience? I want to apply this in my own this way in order to keep the working conditions classroom, but not this year any more. good.

What did you learn? To structure assignments. I know that it is important to structure your lessons; however, this is an element that is always hard for me. It seems easier than it is. Did you plan to learn this, and if so, why did you want to learn this? I did not deliberately plan to concentrate on this, though it is an important issue for me. Structuring assignments often leads to a quieter classroom. Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this? Not relevant. How did you learn this? I learned it by doing and experiencing that some ways do not work. The trial and error idea. Why did you learn it in this way? I did not deliberately choose this strategy, this is how it happens with me. How did you realise that you had learned something? Because I talked with my mentor about ways to improve and also because I had a bad feeling about the assignment. Then you know it is not going well and you have to change. This feeling of need for change is my learning experience. If you look back, about which aspects are you satisfied and what would you do differently next time? I am happy that I became aware of the essence of a good structure, besides I learned from my mentor a new theory about students’ learning processes which I can use to adapt my new strategy in a better way for the students. How will you proceed with this learning experience? I am not making new plans, but I want to make a checklist for myself for criteria for good assignments. In this way I hope to stop making the same mistakes again.

RETROSPECTIVE REGULATION

Figure 3.2. Examples of learning experiences from the four quadrants.

63

0

0

retrospective

-2

prospective

2

-2

-4

passive -4

-2

-4

active 0

retrospective

2

B Active (prospective & retrospective) regulation pattern (N=8)

prospective

A Passive (prospective & retrospective) regulation pattern (N=1)

passive

2

-4

2

prospective

prospective

2

retrospective

-2

-4

passive -4

-2

-4

active 0

retrospective

0

0

-2

active 0

D (Active & Passive) Retrospective regulation pattern (N=2)

C (Active & Passive) Prospective regulation pattern (N=6) 2

-2

passive -4

2

-2

active 0

2

2

prospective

E Versatile regulation pattern (N=5)

-2

retrospective

0

-4

passive -4

-2

active 0

2

Figure 3.3. Typical examples of the five regulation patterns. Every picture shows the positioning of the six learning experiences of a student teacher on the dimensions.

64

Chapter 3

The multiple correspondence analysis positioned all learning experiences on the two dimensions, in one of the quadrants. The six learning experiences of one student teacher formed together a regulation pattern. Five student teachers had their learning experiences spread over three or even all four quadrants. Seventeen student teachers had their learning experience mainly spread over only two quadrants. If at least five out of six learning experiences were positioned in two quadrants (or at least 80% if the student teacher had completed only four or five Learning Reports), then they were categorised in the pattern named after these two quadrants. The five student teachers who could not be classified according to this rule because 80% of their learning experiences were spread over three or four quadrants were categorised as having a versatile regulation pattern. Examples of the pictures of all five regulation patterns are given in Figure 3.3. Table 3.1 Frequencies of regulation patterns among beginner and experienced student teachers Experience of student teachers Regulation patterns

Beginner

Experienced

Total frequency

A

Passive (prospective & retrospective) regulation (quadrants1 & 4)

0

1

1

B

Active (prospective & retrospective) regulation (quadrants 2 & 3)

3

5

8

C

(Active & Passive) Prospective regulation (quadrants 1 & 2)

3

3

6

D

(Active & Passive) Retrospective regulation (quadrants 3 & 4)

2

0

2

E

Versatile regulation (three different quadrants or more)

2

3

5

Missing (fewer than 4 Learning Reports completed)

4

2

6

Total

14

14

28

In Table 3.1 an overview is given of the frequencies of the regulation patterns for the beginner and experienced student teachers separately. One student teacher showed a passive regulation pattern with all learning experiences spread over quadrants 1 and 4. The most frequent regulation pattern in this sample was the active regulation pattern, which combined active prospective and active retrospective regulation (n = 8). Six student teachers combined active and passive prospective regulation in their learning 65

experiences and were therefore categorised as having a prospective regulation pattern. Two student teachers reported learning experiences which were all classified as retrospective. In this small sample, there was no visible trend for differences between beginner and experienced student teachers in their regulation patterns.

3.5.2 Student teachers’ learning conceptions All student teachers were categorised as having a meaning-oriented or a reproduction-oriented learning conception. In total, eight student teachers were categorised as having a meaning-oriented learning conception. Excerpts from the interviews relating to meaning-oriented learning conceptions are given below: A beginner male student teacher of science: “I force myself to think in a systematic way about what I am doing and how I am doing it. To coach myself. Besides, the theoretical background is also important, so also reading about it. That can also give more grip and what I said before, talking and discussing with colleagues.” A beginner female student teacher of mathematics: “So observing, getting feedback and trying things and looking things up in literature to find out alternatives and the background of things. Those are the elements.” An experienced male student teacher of German language: “(…) You have to connect systematically all your knowledge with the feedback you are getting, to get new insights. Yes, I really think, teaching is also being a learner. You really have to keep learning till the final, final end.”

In total, twenty student teachers were categorised as having a reproductionoriented learning conception. Excerpts from the interviews relating to reproduction-oriented learning conceptions are given below: An experienced female student teacher of social science: “It is a process of growth. By doing, by trying, also the experience in your internship, that is very important, then you learn a lot. And then it’s just doing and don’t be afraid, just doing, then you learn a lot.” An experienced female student teacher of art history: “Practising, practice makes perfect. Don’t give up. Of course, you can also read, as background, I do that now and then if I have a problem. But that is not how it works. It is easy of course to see other things [in literature], but still practice rules.” An experienced male student teacher of science: “I would prefer to have a syllabus; that is all I need. That would make a large part of the teacher education programme unnecessary. It is not the same for everybody, but I would prefer a syllabus with a script for every part of the lesson which you have to do.” 66

Chapter 3

In Table 3.2 an overview is given of the frequencies of the learning conceptions for the beginner and experienced student teachers. Of the eight meaning-oriented student teachers, seven were beginners and only one was experienced. This means that although in the beginners' group of student teachers the learning conceptions were evenly spread, in the group of experienced student teachers all except one had a reproduction-oriented learning conception. The difference between the two groups is statistically significant: χ2(1) = 6.30, p = .01. Table 3.2 Frequencies of learning conceptions among beginner and experienced student teachers Experience of student teachers Learning conception

Beginner

Experienced

Total frequency

Reproduction-oriented

7

13

20

Meaning-oriented

7

1

8

Totals

14

14

28

3.5.3 Relations between learning conceptions and regulation patterns In Table 3.3 an overview is given of the patterns of regulation combined with the learning orientations. As stated earlier, an active regulation pattern and a meaning-oriented learning conception as well as a passive regulation pattern and a reproduction-oriented learning conception can be regarded as a theoretically consonant pattern (presented in bold type in Table 3.3). From this perspective, three student teachers showed coherence between their learning conception and regulation pattern; one student teacher combining reproduction-oriented learning with a passive regulation pattern and two student teachers combining a meaning-oriented learning conception with an active regulation pattern. Six student teachers showed a dissonant pattern combining an active regulation pattern with a reproduction-oriented learning conception (presented in italics in Table 3.3).

67

Table 3.3 Frequencies of the different combinations of regulation patterns with learning conceptions1 Learning conception Regulation pattern

Reproductionoriented

Meaning- oriented

A

Passive regulation pattern

1

-

B

Active regulation pattern

6

2

C

Prospective regulation pattern

5

1

D

Retrospective regulation pattern

1

1

E

Versatile regulation pattern

3

2

Missing

4

2

Totals

20

8

Theoretically consonant combinations are presented in bold, theoretically dissonant combinations are presented in italics, and theoretically unclassifiable combinations are underlined. 1

The regulation patterns C, D, and E combine active and passive regulation within the pattern itself (underlined in Table 3.3). This means that it is not possible to describe the combinations of these patterns with a certain learning conception as consonant or dissonant with help of previous theories. In order to give a possible explanation for the existence of these combinations, we will show the data of one student from the most frequent combination among these patterns, namely the combination of a reproduction-oriented learning conception and an active and passive prospective regulation pattern. One of the student teachers with this combination was Maria. Maria was a beginner student teacher of English. She was categorised as having a reproduction-oriented learning conception and, as can be seen in Figure 3.4, three of her six learning experiences were positioned in quadrant 1 (passive prospective), two were positioned in quadrant 2 (active prospective), and one was in quadrant 4 (passive retrospective).

68

2

prospective

Chapter 3

-2

retrospective

0

-4

passive -4

-2

active 0

2

Figure 3.4. Maria’s learning experiences positioned on the two dimensions. Theoretically, the combination of a reproduction-oriented learning conception and a passive regulation pattern could be considered as consonant and such a conception with active regulation as dissonant. Therefore, the two active prospective learning experiences in particular do not fit with her reproduction-oriented learning conception. First we present some quotes from Maria’s interview: (Interviewer) “How do you learn to become a good teacher, could you describe this process? (Maria) I think that you have to observe a lot, also your mentor and your peers, but that you most of all have to do it yourself. I think that that is the solution for everything. (…) (Interviewer) What are the differences with other learning processes? (Maria) When I was studying English it was just theory, this is more practice. The theory is useful as a support now and then, but especially exchanging ideas with your peers and just doing yourself. Learning everything in practice. The book says it nicely, but if you really have to do it, if is of course totally different. (…) (Interviewer) What can you do yourself to become a good teacher? (Maria) Teaching a lot. Trying things. Doing a lot, I think, you have to find your own way of teaching. And you can try to become the same kind of teacher as your mentor, but she is different from me. So, you have to do it yourself”.

69

We can conclude from the interview that Maria had a typical reproduction-oriented learning conception with a focus on learning by doing and keeping what works, without valuing and using theory or other interpretations for evaluating teaching experiences. On the other hand, she also reported two learning experiences that were active prospectively regulated. The two active prospective learning experiences are described in Figure 3.5.

EXAMPLE 1 What did you learn? I planned a lesson for my first-year students and I wanted to give that lesson exactly how I prepared it. But it went wrong. Did you plan to learn this, and if so, why did you want to learn this? Yes, because it is always nice if you can give a lesson how you prepared it. Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this? Yes, because until now I always succeeded in following my plans. How did you learn this? I planned to start with a new chapter, because we had just finished the last one, but I did not take into account that not every moment is a good moment to start a new chapter (it was the last day before the spring holiday). Why did you learn it in this way? [No answer] How did you realise that you had learned something? My lesson did not work out the way I planned it. I realised immediately that this was not going to work. The students were very excited (…) If you look back, about which aspects are you satisfied and what would you do differently next time? I just learned an important lesson. Next time, I will plan my last lesson before a holiday totally differently. How will you proceed with this learning experience? See above.

EXAMPLE 2 What did you learn? I learned that personal attention is very important and often works better than corrections. An example: a student does not start with the assignment. Instead of correcting him in front of the classroom, I approach him personally and ask him to start. Did you plan to learn this, and if so, why did you want to learn this? Yes, I noticed that correcting often worked in the opposite way and I wanted to find a way to get everybody’s attention on the assignment. Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this? Actually not, it started as a test, but I immediately noticed that it worked. How did you learn this? The student didn’t want to start working and I approached him personally to tell him to start working. After that he started working. Why did you learn it in this way? Correcting did not work any more so I wanted to try other things. This was the first try. How did you realise that you had learned something? The boy did start to work. If you look back, about which aspects are you satisfied and what would you do differently next time? I am satisfied with everything, I just wish that I had done this before, but I am proud that I initiated this myself. How will you proceed with this learning experience? I am going to do it more often.

Figure 3.5. Description of the two active prospective learning experiences of Maria. These learning experiences were both planned, and Maria thought seriously about why she wanted to learn it, her strategy and whether she was going to succeed in it. The regulation activities were to a large extent focused on steering her teaching behaviour in order to learn from it. Furthermore, coherent with the positioning in the active prospective regulation quadrant, the examples did not show evidence of active retrospective regulation as well. The reflection on what is learned was quite superficial: she knew that she had learned something, because she experienced that it worked, and in her evaluation she was almost completely satisfied. This example shows that it is possible to plan and set up experiences to learn from in a sense of step by step improving teaching behaviour without questioning the interpretative 70

Chapter 3

framework through critical reflection or evaluation. In this case, for an active prospective regulation of a learning experience (planning, choosing a strategy, self-efficacy notions) it was not necessary to have a meaning-oriented learning conception.

3.6 Discussion and conclusion The aim of this study was to increase our understanding of student teachers’ self-regulated learning in a dual teacher education programme. We looked at both the more dynamic regulation component of SRL (self-regulation activities) and the more stable component (conceptions of learning), and the relation between them. Learning Reports with open questions were developed for this study to measure student teachers’ regulation activities in different learning experiences. Multiple correspondence analysis of the data showed that every learning experience could be positioned on two dimensions that were interpreted as discerning passive and active regulation and prospective and retrospective regulation. In this study, the dimension retrospective versus prospective regulation of learning turned out to be an important dimension in which the student teachers’ learning experiences differed. This is an addition to previous research on self-regulated learning, where often only the dimension self versus external regulation has been found (Vermunt & Vermetten, 2004). This can be explained by the significance of unplanned learning experiences in student teachers’ development. The six learning experiences of every student teacher were pictured in a structure of two dimensions in one of the quadrants in order to describe their individual regulation pattern. For all student teachers, variation was found among their six learning experiences in terms of how these were regulated. Five student teachers were categorised as having a versatile regulation pattern and one of the student teachers had a clearly passive regulation pattern. The rest of the student teachers could be categorised as being active, prospective or retrospective regulators. Since the prospective and retrospective regulation patterns consisted both actively and passively regulated learning experiences, almost all of the student teachers in our study were able to regulate their learning at some moment in some situation in an active way. No trend was visible in this sample for differences between beginner and experienced student teachers in their regulation patterns.

71

The question remains why there is so much variation within a student teacher’s regulation pattern. The result that some student teachers reported both prospectively and retrospectively regulated learning experiences can be explained by the fact that in a dual learning programme student teachers learn from planned as well from more spontaneous learning experiences. In the case of a spontaneous learning experience it is not possible to have an actively regulated forethought phase, but it is possible to get the most out of the learning experience by regulating it retrospectively. For the combination of active and passive learning experiences within one student teacher different explanations may be possible. It may be the case that these student teachers are in a change phase of their learning approach, which often occurs on entry to a new environment (Boulton-Lewis, Wilss, & Lewis, 2003; Brownlee et al., 2003; Cano, 2005; Lindblom-Ylänne & Lonka, 2000): student teachers may be capable of active regulation in some situations, but not (yet) in all. Furthermore, in learning experiences at the practice school, there might be a tension between the inward focus on the development of self-as-teacher and the outward focus on the concerns about their students, as described by Conway and Clarke (2003). It might be a deliberate choice of a student teacher to focus in some situations primarily on the well-being and learning of their students and to pay less attention to their own learning, resulting in a more passive regulated learning experience. We could therefore also ask the question whether it is actually necessary and desirable for student teachers to regulate every learning experience in an active way. It could be seen as a characteristic of a dual learning programme that student teachers show a combination of multiple types of regulation. More research is needed to interpret whether passive regulation is caused by a lack of capacity, or by a deliberate or non-deliberate choice of the learner in a certain learning experience. The learning conceptions of all student teachers were categorised as being meaning-oriented or reproduction-oriented in nature. In our sample, only a small number of student teachers (eight, i.e. 29%) showed a meaningoriented learning conception. Between the groups of beginner and experienced student teachers there was a significant difference in the frequencies of a meaning-oriented learning conception; only one student teacher’s learning conception from the group of fourteen experienced student teachers was categorised as meaning-oriented. Since we did not follow the same student teachers during the teacher education programme, we cannot draw any conclusions about the development of these individual student teachers. 72

Chapter 3

However, the results are not in line with the expectation of Oosterheert (2001) that student teachers would change towards more meaning-oriented learning throughout the programme. A possible explanation may be that experienced student teachers have to give many more lessons and have greater responsibility as teachers at the end of the programme. This may lead to a stronger focus on practice as a main source for learning and less time for questioning and re-interpreting interpretative framework. Nevertheless, since we know that a meaning-oriented learning approach is beneficial for learning (Lindblom-Ylänne & Lonka, 1999; Vermunt & Vermetten, 2004), it is an intriguing question why so many student teachers who are supposed to become experts in learning have a reproduction-oriented conception. Further longitudinal research is necessary to study the development of student teachers’ learning conceptions in more detail. In this study we used the terms consonance and dissonance to look at the relation between regulation patterns and learning conceptions of individual student teachers. From a theoretical perspective, only three of the 22 student teachers showed a fully consonant combination between regulation of learning and learning conception and six student teachers showed a dissonant combination. This dissonant combination does not necessarily have to be interpreted as negative, since dissonance may also be an important condition for change and development (Vermunt & Minnaert, 2003). The other thirteen student teachers showed examples of both active and passive regulation. Therefore, their combinations of regulation patterns and learning conceptions could not be classified as clearly consonant of dissonant. Detailed analysis of one of these student teachers showed, however, that a reproduction-oriented learning conception could go very well together with an active prospective regulation pattern in which student teachers try to improve their learning step by step without questioning their framework of reference. Although it has often been stated that for a meaning-oriented learning approach, active regulation is needed (e.g., Biggs, 1988), our data showed that the opposite does not have to be the case. Also, previous research by Brownlee et al. (2003) has shown that deep or meaning-oriented learning is often characterised complex interactions of deep and surface strategies, with an overall intention to develop personal meaning. This may explain the combination of a meaning-oriented learning conception with a regulation pattern consisting of active and passive regulated learning experiences.

73

Altogether, we have to conclude that the relation between student teachers’ learning conception and regulation pattern is more complex than simply relating meaning-oriented learning to active regulation and reproduction-oriented learning to more passive regulation. Further research with a larger sample is necessary to arrive at a more detailed theory on the relationship between different learning conceptions and their regulation patterns and the meaning of consonance and dissonance in this respect. In this study, we only focused on the concept of learning conception and for this small sample only a rough distinction between meaning-oriented and reproduction-oriented learning was used for categorisation. In a larger sample, more detailed descriptions of student teachers’ stable part of SRL might be obtained by studying student teachers’ general learning orientations, in which also their preferences for learning and regulation activities are included (Oosterheert, Vermunt, & Denessen, 2002). Since the data for this study have been collected at one teacher education institute in a cross-sectional way, more research is needed to confirm our results concerning the different regulation patterns and to give more insight into student teachers’ development of learning and regulation. In this study, we have shown the benefits of an instrument such as the Learning Report to expose the variation in regulation activities within student teachers. Dissonant patterns and variation in learning across learning environments are much harder to unravel with general questionnaires about student teachers’ learning (Lindblom-Ylänne, 2003). Nevertheless, these instruments are more practical in use for large-scale and longitudinal studies. Therefore, future research has to find ways to upscale qualitative instruments as the Learning Report or to seek different methods of analysis for existing questionnaires in order to capture and value the variation within persons over time and across learning experiences. In today’s teacher education institutes as well as in other forms of vocational education, self-regulated learning is highly valued. Students are expected to develop themselves in an active, reflective and deliberate way. They are obliged to make personal development plans including learning goals, learning strategies, time schedule, and an evaluation of whether they have succeeded in reaching their preset goals. In this way, these tools focus especially on stimulating the prospective side of regulation. As we have seen in our data, a large number of student teachers’ learning experiences is unplanned. In these unplanned learning experiences, which can be of great 74

Chapter 3

value in student teachers’ development, active retrospective regulation is also necessary. We believe a pedagogy should be developed for teacher education that stimulates student teachers to use both prospective and retrospective regulated learning to become active owners of their development as a teacher.

75

discriminate

information traditional

teaching

myself

active

usability

weeks

calculated

comparable

amount

consistency

asked

characteristics

liked

argumentation

relation varied

options experienced procedure following perry student teachers' monitoring iltp srl programme prospective self-regulated separated report validity focused

extent

aim

regulation

besides

unplanned

learner

appendix off-line classified happened

evaluation

situation

contexts distinction

namely

discussed

portfolio

corno

existing

competences

research

independent

vermunt coded obtain

choice conclusion

framework

qualities

assessment

dual

specific

multiple

meaning-oriented

students

beforehand items

something relevant

dimensional

analyzed

included

dimension

diaries

construct

reflection

strategies

collection

education

learning report

outcomes

categories

recognised

phase

personal

correspondence

reliability rate

content

respectively

sample aspects

suggested

significant

orientation

interpreted

inventory

aggregated extra formulation categorised winne versus retrospective methods developed receivedresults otherwise underlying on-line wolters combined missing planned previous completed problems actual internal task selection

experiences

available

considering

analysis data

indications goal

concurrent

correlations

learning

succeed multiple-event

freq

rubrics

variables predictive

variation

kinds

suitable

positive teacher terms satisfactory aptitude scores wanted boekaerts

types version

explained

inter-rater

psychometric

participants examples

structured

quadrants

variances

criteria measurement cascallar

knowledge

description lack questionnaire pintrich

occasions

passive

university ability

concept

described self-reports

different

qualitative

across

making

means

4 The development and evaluation of an instrument to measure student teachers’ regulation of learning in dual learning programmes4 Abstract Two studies were conducted to develop and evaluate the Structured Learning Report: a multiple-event instrument to measure student teachers’ regulation of learning in a dual learning programme. In Study 1, data was collected from 28 student teachers with an open question log to collect qualitatively rich descriptions of student teachers’ variation in regulation activities from both contexts of the dual learning programme. In Study 2, these descriptions were used to develop the Structured Learning Report, which has been examined in a new sample of 90 student teachers. Aggregation of multiple events with the help of multiple correspondence analysis made it possible to discriminate different qualities of student teachers’ regulation, without relying on the student teachers’ ability to rate themselves across learning experiences. Although construct validity and reliability were satisfactory, no significant relations were found with student teachers’ learning outcomes and general learning orientations. More measurements with larger samples are needed to determine the predictive and concurrent validity of the instrument.

This chapter has been submitted in adapted form as: Endedijk, M., Brekelmans, M., Den Brok, P., & Vermunt, J.D. Development and evaluation of an instrument to measure student teachers’ regulation of learning in dual learning programmes. 4

77

4.1 Introduction In recent decades research on self-regulated learning (SRL) has increased enormously. Different models have been developed to describe the: “active, constructive process whereby learners set goals for their learning and attempt to monitor, regulate and control their cognition, motivation, and behaviour, guided and constrained by their goals and the contextual features in the environment” (Pintrich, 2000, p. 453). SRL has benefits for academic performance (Cantwell & Moore, 1996; Vermunt, 2005), but also expertise development is dependent on professionals’ ability to actively create and regulate their own learning experiences (Ericsson, 2006). Therefore, not only academic learning programmes, but also different kinds of vocational education put more demands on students (to learn) to regulate their own learning. Dual teacher education programmes are one of the domains in which students are often required to self-regulate their learning to a high level: in these programmes student teachers combine workplace learning with studying at university. Since they learn in different learning contexts, they also have more freedom in shaping their own curriculum by making their own professional development plans, creating learning experiences, monitoring their learning and reflecting in a portfolio. According to Boekaerts and Cascallar (2006), accurate information regarding three issues is essential to support learners in their development as self-regulated learners. First, it is essential to have a clear description of the self-regulation strategies that students need to steer their learning in a specific domain. Second, valid assessment instruments are required to assess the ease with which students use these strategies. Finally, information must be obtained as to how aware students are of the self-regulation strategies they are currently using (Boekaerts & Cascallar, 2006). Past research into SRL has mainly focused on traditional instructional programmes in secondary or higher education, most models have been developed for school learning (Corno, 1986) and instruments have also been developed for these contexts. Unfortunately, these instruments are unsuitable for dual learning programmes, because of their specific focus on task-centred learning or studying from books. Since student teachers in dual learning programmes often create their own learning paths in varying contexts, their learning is hardly organised around concrete, small tasks. Therefore, an instrument to measure the regulation activities is urgently needed for this type of complex learning environment. The application of such an instrument in teacher education programmes might provide information to 78

Chapter 4

teacher educators about specific student teachers’ needs to develop their regulation of learning ability. In the theoretical framework we will discuss the concept of SRL and various types of instruments that have been developed so far, to come to a set of criteria for an instrument for the context of a dual teacher education programme.

4.2 Theoretical framework 4.2.1 Defining and measuring self-regulated learning There is considerable agreement about the importance of self-regulated learning (SRL), but there has been disagreement about how it can be conceptualised and operationalised in a scientifically useful way (Alexander, 2008; Boekaerts & Corno, 2005; Zimmerman, 2000). The review of Boekaerts and Corno (2005) shows how different definitions of SRL also resulted in differences in the way it has been assessed. Initially, the concept of SRL has been viewed as a stable individual characteristic resulting in de-contextualised trait-like measurements. The situated learning approach has resulted in a view of SRL as more a dynamic context-dependent set of activities, so instruments were developed to measure SRL in real time (Boekaerts & Cascallar, 2006; Boekaerts & Corno, 2005). This also lead to the development of more qualitative and ecologically valid instruments (Butler, 2002; Cascallar, Boekaerts, & Costigan, 2006; Perry, 2002). In current instruments, the different operationalisations of the concept of SRL can still be recognised. Winne and Perry (2000) made a distinction between instruments that measure SRL as an aptitude and instruments that measure SRL as an event. An event-instrument describes the regulation activities during a specific task. When SRL is measured as an aptitude, a single measurement is used to identify a relative enduring attribute of a person (Winne & Perry, 2000). Van Hout Wolters (2000) divided instruments into online and off-line methods. This distinction is related to the moment SRL is measured. On-line methods measure SRL during the learning task, off-line methods measure SRL independently from or directly after a learning task (Van Hout Wolters, 2000). This last distinction is sometimes seen as overlapping with the aptitude-event measurement distinction. Although aptitude instruments are always used off-line, there are also examples of offline event-measurement. In Table 4.1, we classified the types of instruments mentioned in several overviews (Boekaerts & Corno, 2005; Van Hout Wolters, 79

2000, 2006; Van Hout Wolters, Simons, & Volet, 2000; Winne & Perry, 2000) according to these distinctions. Table 4.1 Classification of the different types of instruments to measure SRL On-line Aptitude

Off-line • • •

Event

• • • •



Think-aloud methods Eye-movement registration Observation and videoregistration of behaviour Performance assessment through concrete study tasks, situational manipulations or error detection tasks Trace analysis

• • • •

General self-report questionnaires General oral interviews General teacher judgments Stimulated recall interviews Portfolios and diaries/ logs Task-based questionnaire or interview Hypothetical task interview

As can be seen in Table 4.1 there are different types of instruments available to measure SRL. There has been debate concerning the pros and cons of the different types of instruments mentioned. A review study by Dinsmore, Alexander, and Loughlin (2008) showed, that from the 75 studies included that measured SRL, 59% measured it by means of decontextualised self-reports. This strong reliance on aptitude instruments has often come under fire (Dinsmore et al., 2008). The central argument against these instruments is that it is not clear which situations the learners have in mind and which references they have for comparison when completing these questionnaires (Van Hout Wolters, 2000). Although these instruments can be reliable and valid, many authors consider the results to be poor indicators of the actual regulation activities that students use while studying (e.g., Perry, 2002; Perry & Winne, 2006; Pintrich, 2004; Veenman, 2005; Winne & Perry, 2000). This often leads to low predicative values of these instruments for learning outcomes and low correlations with on-line methods (Veenman, 2005). However, self-report instruments such as the Motivated Strategies for Learning Questionnaire (MSLQ) (Pintrich & Smith, 1993), Inventory of Learning Styles (ILS) (Vermunt, 1998), Metacognitive Awareness Inventory (MAI) (Schraw & Dennison, 1994), and Learning and Study Strategies Inventory (LASSI) (Weinstein, Schulte, & Palmer, 1987) are valuable for measuring what students perceive to be their 80

Chapter 4

general learning preferences, as well as their general motivation and capacity for self-regulation (Perry & Winne, 2006; Pintrich, 2004; Zimmerman, 2008). These instruments are also most practical in use (Pintrich, 2004). For the context of teacher education, Oosterheert, Vermunt, and Denessen (2002) developed an aptitude instrument, namely the Inventory Learning to Teach Process (ILTP). This instrument measures student teachers’ learning orientation in which preferences for learning and regulation activities have also been included. The alternative approach of aptitude measurement is to measure SRL as an event, during an experience or task that is marked by a prior and following event (Winne & Perry, 2000). These latter instruments are more suitable for finding relations between specific aspects of real time SRL in authentic contexts (Zimmerman, 2008). The advantage of on-line methods is that the measurement is at the same moment as the task: little information about what happens during the task is lost (Van Hout Wolters, 2006). Despite these benefits, with on-line methods there is a bigger chance of influencing the learning process by for example prompting students to think aloud (Greene & Azevedo, 2009). Furthermore, on-line methods only take into account the SRL activities which are performed during the observed learning activity. Moreover, to measure SRL on-line it is essential to have the instrument present during the task. Therefore, for contexts of workplace learning in which students do not learn with the help of pre-set tasks and in which learning is often unplanned (Tynjälä, 2008), on-line measurement of SRL is not an option. The off-line event measurement of SRL has less frequently been discussed. Compared to on-line event methods, these instruments have the capacity of also measuring more tacit aspects of SRL for which the students need some time to recollect what exactly happened during an experience (Howard-Rose & Winne, 1993). Portfolios and diaries have been suggested as having the potential to measure SRL in a reliable and valid way, but until now they have mostly been used as an intervention to stimulate students’ SRL (Meeus, Van Petegem, & Engels, 2009; Zimmerman, 2008). Diaries have been shown to be equal or more sensitive than pre- and post-test questionnaires when it comes to measuring changes in SRL in ecologically valid contexts (Zimmerman, 2008). There is, however, a standardization problem with these instruments. Since every portfolio or set of diaries consists of a unique set of products, students are assessed on the basis of different learning experiences (Van Hout Wolters, 2000). From studies on experienced teachers, we know that 81

a digital diary is a suitable instrument to collect different kinds of learning experiences (Bakkenes, Vermunt, & Wubbels, in press; Hoekstra, Brekelmans, Beijaard, & Korthagen, 2009; Meirink, Meijer, Verloop, & Bergen, 2009; Van Eekelen, Boshuizen, & Vermunt, 2005; Zwart, Wubbels, Bolhuis, & Bergen, 2008). In conclusion, both aptitude and event measurements have the potential to contribute to our understanding of students’ SRL (Howard-Rose & Winne, 1993; Winne & Perry, 2000). This means that no single instrument is capable of capturing all aspects of students’ SRL (Boekaerts & Cascallar, 2006). The choice of instruments depends on the nature of the research problem and the context (Boekaerts & Cascallar, 2006; Cascallar et al., 2006; Lonka, Olkinuora, & Mäkinen, 2004; Pintrich, 2004). Besides the aptitude instrument (ILTP) developed by Oosterheert, Vermunt, and Denessen (2002), to our knowledge no instrument is available that can measure student teachers’ SRL in a dual learning programme. For the measurement of student teachers’ regulation of concrete learning experiences in both contexts of a dual teacher education programme, an instrument has to be developed. The characteristics of the different types of instruments as described above suggest that the most suitable type of instrument for this purpose is an off-line event instrument. Besides the choice for a specific type of instrument, Van Hout-Wolters (2000) has listed a number of aspects which are important in the selection of an appropriate method for assessing learning skills. These include: the goal of the assessment, the type of data to be collected, the way of data processing, the financial aspects of the data collection, the content of the assessment (which skills are assessed), the participants and context, the assessment procedure, and the psychometric quality of the instrument. Below, these aspects will be discussed for the context of a programme in which studying at university and workplace learning are combined.

4.2.2 Towards criteria for the instrument The goal of the assessment is to diagnose student teachers’ quality of regulation across different learning experiences from both contexts of the dual learning programme. Eventually, the instrument should be practical enough to be used on a large scale for repeated measures, to be able to diagnose all student teachers’ quality of regulation during various moments of the teacher education programme. This has consequences for the type of data to be collected, the processing of the data and the financial aspects of the data collection. 82

Chapter 4

The content of the assessment will be the actual regulation activities that student teachers use in learning to teach. The participants are student teachers who learn in both contexts of a dual learning programme: at universities and at schools. As described above, since the curriculum does not consist of fixed tasks, there is a large variability in student teachers’ learning experiences. This means that the instrument should be able to cover different kinds of learning experiences (e.g., planned and unplanned), in different contexts, with varying duration. This has consequences for the assessment procedure. We have already concluded that an off-line event instrument will be most suitable. Moreover, the variation in learning experiences makes it necessary to include multiple learning experiences to give a reliable estimation of student teachers’ quality of regulation. Finally, the psychometric quality of the instrument should be high enough to discriminate different qualities of student teachers’ regulation of learning in a reliable and valid way. In summary, the following criteria can be set for the instrument: it should measure off-line, in a reliable and valid way, student teachers’ regulation activities during multiple and different kinds of learning events from the two contexts of the dual learning programme. Aggregation of these multiple events should discriminate different qualities of student teachers’ regulation of learning. From the different off-line event measurements listed in Table 4.1, the hypothetical and stimulated-recall interview do not meet these criteria, since they are too labour intensive to use for multiple-event measurements with a relatively large number of participants. Although the portfolio has been suggested as a valid instrument, the use of portfolios varies a lot among student teachers and teacher educators and is therefore in itself not structured enough to collect comparable data of all aspects of the regulation process (Van Tartwijk, Driessen, Van der Vleuten, & Stokking, 2007). As mentioned previously, the diary or log has been used successfully before to collect different kinds of teachers’ learning experiences and seems therefore to be the most suitable instrument for this context. The main research question of this study is: To what extent is a multiple event question log a valid and reliable instrument for measuring and discriminating different qualities of student teachers’ regulation of learning in a dual learning programme? We used a two-step procedure to answer this question. As Cascallar et al. (2006) mentioned, the first step is to have a clear description of the relevant regulation activities that are necessary for students to steer their learning in a specific domain. Since this description is not yet 83

available, the first step was to develop an open question log, aimed at collecting qualitatively rich descriptions of student teachers’ variation in regulation activities across different learning experiences. The qualitative data generated by this instrument have been used to develop in the second step a structured question log that is less labour intensive, but still meets criteria of reliability, validity, and discriminative power.

4.3 Study 1: Examination of the Open Question Learning Report 4.3.1 Instrument For the measurement of qualitative differences in student teachers’ regulation activities an open question log, called the Learning Report, was developed. Based on the commonalities of different SRL models, Pintrich (2000; 2004) developed a conceptual framework that was used in this study for the development of the question log. This framework classifies different phases of regulation for the assessment of SRL in the college classroom. The four phases reflect goal-setting, monitoring, control, and regulation processes and the areas cover cognition, affect, behaviour, and context. For every phase of SRL as described by Pintrich (2000) two or three questions were included in the Learning Report (see Appendix A): for the planning and goal-setting phase questions were inserted about student teachers’ goal orientation (Question 2), notions of self-efficacy (Question 3), learning strategy (Question 4) and forethought (Question 8). The questions concerning the monitoring and control phase described their control of the learning strategy (Question 5) and monitoring of the learning results (Question 6). The questions concerning the reflection phase were focused on the learning outcome (Question 1) and evaluation of the learning experience (Question 7). Phases 2 (monitoring) and 3 (control/regulation) were taken together, because between these two processes not much differentiation has been found with off-line methods (Pintrich, 2000). To collect information from multiple events, the whole instrument consisted of six Learning Reports in which student teachers could report their six selfchosen learning experiences. A pilot test with three student teachers was completed to check their comprehension of the questions; some small adaptations were made.

84

Chapter 4

4.3.2 Method study 1 Twenty-eight students from a Dutch dual teacher education programme volunteered to participate in this study. The student teachers were selected in a stratified sample intended to capture as much variation as possible in the following variables: teaching experience, supervising teacher educator, school subject, experience and gender. A total of 133 Learning Reports were collected during a period of six weeks. Eighteen student teachers completed all six Learning Reports as required, five completed only a part. Phenomenographically informed content analysis of the 133 Learning Reports showed that the variation within each of the eight questions could be described by five to seven categories. All text fragments of the Learning Report were coded in terms of the set of categories developed. To discover relationships between categorical data and the underlying structure in the data, multiple correspondence analysis (Mair & De Leeuw, 2008) was used. This is a form of nonlinear principal component analysis (De Leeuw, 1982). The outcome of the multiple correspondence analysis showed that the large variation in student teachers’ regulation activities could be described in terms of an underlying structure of two dimensions. Dimensions 1 and 2 explained respectively 45.1% and 33.9% of the variance in the data. The first dimension discerned passive regulation from active regulation of learning. Learning experiences that were passively regulated by the student teachers were characterised by a lack of argumentation for decisions they had made as well as answers that showed that someone else was in charge of the learning process. Furthermore, many aspects of the regulation process were not described at all. An actively regulated learning experience showed a more active pattern of regulation where the student teacher reflected more deeply on the learning content, the learning process and his or her own role in it. These student teachers made their own decisions for a learning strategy, but also actively used information from others and used it in their reflections. The second dimension separated prospective and retrospective regulation of learning from each other. In a prospectively regulated learning experience, the focus of the regulation activities is on the first phase of the learning process. The learning experience was planned, goals were set and an argumentation for choosing a learning strategy was given. The phase after the learning experience received less attention; the monitoring, reflection, and evaluation were more superficial. The retrospectively regulated learning experiences were often unplanned, so no goal-setting or deliberate thinking 85

about learning strategy and self-efficacy had taken place. The regulation focused on the monitoring, evaluation and reflection part of the learning process. The positioning of all learning experiences on the dimensions is pictured in Figure 4.1. Every learning experience received a score on these two dimensions that described its activeness or passiveness and whether the focus of the regulation was prospective or more retrospective. For every student teacher, a regulation pattern was established based on the combination of scores of their six Learning Reports (or fewer if they had not submitted all the Learning Reports) on the two dimensions.

2

-2

ACTIVE REGULATION

0

PASSIVE REGULATION

PROSPECTIVE REGULATION

-4

-6

RETROSPECTIVE REGULATION

-8 -4

-2

0

2

Figure 4.1. Positioning of all 133 learning experiences from Study 1 on the two dimensions.

4.3.3 Results study 1 Reliability The reliability of the instrument can be calculated for the categorisation of the qualitative data as well as for the internal consistency of the instrument itself. From the total of 1197 fragments that were coded, 10% were also coded by an independent second researcher to compute inter-rater reliability. After a small 86

Chapter 4

adaption of the initial codes, an overall Cohen’s Kappa of the eight variables of .90 was reached. The internal consistency of the two dimensions could be calculated with Cronbach’s alpha, comparable with how Cronbach’s alpha is used to estimate the reliability of a scale in a questionnaire. In this case, however, the reliability of the dimensions is calculated based on students’ scores on the eight variables. Both dimensions showed a satisfactory internal consistency: dimension 1, passive versus active regulation, Cronbach’s X = .83; dimension 2, retrospective versus prospective regulation, Cronbach’s X = .72.

Validity In this study construct validity was assessed by comparing the structure in the data with structures found in other research. The amount of explained variance as described above shows that the dimensions contribute well to the explanation of variances in the eight variables. All phases of regulation as described by Pintrich (2000), were covered by the questions of the open question log. Comparable with previous studies, one of the dimensions underlying the data showed how actively a learning experience is regulated. Although in previous research this dimension was often named self-regulation versus external regulation (Kaplan, 2008; Vermunt, 1998), this definition turned out to be less relevant for the present data. In our data set, external and lack of regulation was found at the same side of the dimension. On the other side of the dimension, examples of regulation by the student teacher were found. Therefore, we interpreted this dimension as active versus passive regulation, in which passive regulation included external as well as lack of regulation. The second dimension, retrospective versus prospective regulation, is a dimension that to our knowledge has not been identified before in the literature. The existence of this dimension can be explained by the significance of unplanned learning experiences in student teachers’ development (Hodkinson & Hodkinson, 2005). Previous studies into SRL have stressed the crucial role of planning in the active regulation of learning (Eilam & Aharon, 2003). However, this dimension also shows how unplanned learning experiences can still be regulated in an active way, although more retrospectively.

Discriminative power The instrument measured different types of learning experiences in different contexts: all student teachers reported learning experiences from learning at 87

university as well as learning from practice at school. Furthermore, from the 133 learning experiences, 50 experiences were unintentional and 83 were said to be planned. Among the 22 student teachers who submitted enough Learning Reports to be included in the analysis, five different regulation patterns could be distinguished based on the quadrants in which their experiences were located. Five student teachers had their six learning experiences spread over three or even all quadrants of Figure 4.1. These student teachers were classified as having a versatile regulation pattern. Seventeen student teachers had their learning experience spread over only two quadrants. One student teacher showed a passive regulation pattern with all learning experiences spread over the passive side of the first dimension. The most frequent regulation pattern in this sample was the active regulation pattern, which combines active prospective and active retrospective regulation (n = 8). Six student teachers combined active and passive prospective regulation in their learning experiences and were therefore categorised as having a prospective regulation pattern. Two student teachers were all classified as having a retrospective regulation pattern.

4.3.4 Conclusion study 1 The Open Question Learning Report generated qualitative descriptions of the large variations in student teachers’ regulation activities. Multiple correspondence analysis suggested two underlying dimensions, which described the activeness of student teachers’ learning experiences and whether they regulated their learning more prospectively or retrospectively. Inter-rater reliability, internal consistency and construct validity were satisfactory. By using the Learning Report as a multiple-event instrument, different qualities of student teachers’ regulation of learning by means of regulation patterns could also be discriminated. Therefore, it was concluded that the Open Question Learning Report could be used for the development of a structured, less labour-intensive version.

4.4 Study 2: Development and evaluation of the Structured Learning Report The aim of the second study was twofold: First, to replicate the dimensional structure that was found with the open question log in a new sample with a structured, less labour-intensive version of the instrument. Second, to obtain 88

Chapter 4

indications of predictive validity, concurrent validity and usability of the instrument for teacher education.

4.4.1 Method study 2 The Structured Learning Report This instrument was developed on the basis of the results of the Open Question Learning Report as described previously. The same eight variables, each reflecting separate aspects of self-regulation were phrased in eight questions, but this time multiple choice items were used. The options of the multiple choice questions reflected the categories from the content analysis of the Open Question Learning Report (see Appendix B). For the formulations of the options, we made use of the examples from the data from the Open Question Learning Report. A pilot test with four student teachers was completed to check their understanding of the questions and multiple-choice options. This resulted in some small adaptations of the formulation of the options. The number of choices per question varied between five and eight. Besides, for every question there was also the opportunity to use the option “otherwise, namely…”, allowing participants to describe the answer in their own words in case it did not fit the multiple choice options. The Structured Learning Report started with one open question: What did you learn? In this way, the student teachers were able to describe in their own words their reflections on their object of learning. The same categorisation of reflections was found as in the previous study. These reflections were categorised by the researcher according to the coding scheme. The Structured Learning Report was transformed into a web-based questionnaire which could be accessed at any moment during the data collection period by the student teacher. Since some of the questions were not relevant for unplanned learning experiences, this online tool made it possible to follow special routes to skip questions in case they were not relevant as in the case of unplanned learning experiences. Again the whole instrument consisted of six Learning Reports. Extra measurements Extra measurements were completed to collect first indications for the predictive validity, concurrent validity and usability for teacher education. From previous research we know that there is a positive relationship between active regulation and learning outcomes (Cantwell & Moore, 1996; Pintrich & De Groot, 1990; Vermunt, 2005). To determine predictive validity, standardised 89

tests would be needed. At the teacher education institute where the data was collected, the general level of competence of student teachers was assessed by making use of portfolios. Based on the portfolio and a final interview with the supervising teacher educator, the student teacher gains his qualification or has to do an extra term. The content of the portfolio consists of all products that the student teacher has made (e.g., lesson plans), reflections on their learning, a professional development plan, and feedback from their supervisors and pupils. No grades or other quantitative assessments are given. During our data collection period, scoring rubrics (Jonsson & Svingby, 2007) were introduced to assist teacher educators with the assessment of student teachers’ teaching competences. With help of these rubrics, the teacher educators could rate the student teacher on six different competences representing the teacher’s profession on a scale from one to four, with four representing excellent teaching. For our study, we calculated the student teachers’ final assessment score by computing the mean of the teacher educator’s ratings on the six competences at the end of the programme. An indication for predictive validity of the Structured Learning Report was calculated by relating the final assessment score based on the rubrics to the student teachers’ scores on the regulation dimensions quality of regulation student teachers shown at the end of the programme. Our hypothesis is, in line with previous studies, that active regulation is related to a higher final assessment score at the end of the programme. An indication of concurrent validity was collected by making use of the ILTP, developed by Oosterheert, Vermunt, and Denessen (2002). This aptitude instrument with Likert-scale items measures student teachers’ learning orientation and categorises students into four learning orientations: survival, reproduction, closed meaning, and open meaning-oriented. Open meaningoriented learning is distinguished from closed meaning-oriented learning by the independency of the student teachers in extending their frame of reference to develop themselves as a teacher. Although previous research has often shown very low correlations between self-report questionnaires and eventmeasurement, at least we can expect that if there is a relation, this will be in the expected direction. In line with previous research, a positive relation between active regulation of learning and a meaning-oriented learning orientation, as well as a positive relation between survival and reproduction-oriented learning and passive regulation can be expected (Beishuizen, Stoutjesdijk, & Van Putten, 1994; Vermunt, 1998). Since the dimension prospective90

Chapter 4

retrospective has not been found in previous studies, no hypotheses for this dimension can be set. To obtain more information considering the usability of the instrument for teacher education, an evaluation questionnaire was developed to collect information of how the student teachers had perceived their participation in this study and the use of the instruments. The question asked on a Likert-scale from one (absolutely not) to seven (totally), to what extent they recognised the outcomes of the Structured Learning Report and ILTP, which were returned to them in a personal report of their development as a learner. Student teachers were given the opportunity to provide comments in an open question.

Participants All 90 full-time student teachers from one post-graduate teacher education institute in The Netherlands were asked to volunteer in this study. All student teachers were initially willing to participate; however, five student teachers cancelled their participation because of a lack of time, illness, pregnancy, other expectations of the study, and one left the teacher education programme. The total set of participants consisted of 22 male and 63 female student teachers teaching one of the 18 different general (i.e., non-vocational) secondary school subjects.

Procedure The data collection period consisted of three periods of two weeks in which student teachers were instructed to report six different learning experiences during every period online in the Structured Learning Report. The periods of data collection took place three (T1), six (T2), and nine months (T3) after the start of their programme. After every period of two weeks reminders were sent to collect the missing Learning Reports. The student teachers were asked to choose a learning experience that occurred a maximum of two weeks earlier; this could be any kind of experience that was part of their development as a teacher. They were stimulated to describe different kinds of learning experiences in the six Learning Reports: two learning experiences that had taken place (at least partially) in the context of the teacher education institute, two that had taken place (at least partially) in the context of the practice school, and two of their own free choice. Furthermore, they were instructed that they could report planned as well as spontaneous learning experiences, and both successful and unsuccessful learning experiences. In the same three periods they were also asked to complete the ILTP. As a reward they received a 91

personal description of their development as a learner which they could use as evidence in their portfolio. After the personal description was sent, they received a link to the online evaluation questionnaire. In total, 66 student teachers completed all 18 Learning Reports, five student teachers missed one data collection period and six student teachers participated only during one measurement occasion. Four student teachers completed less than four Learning Reports, and were therefore excluded from the analyses. Four student teachers did not complete a single Learning Report during the three data collection periods but provided no reasons for this. This resulted in 75 participants during the first period, 71 participants during the second period, and 69 participants during the last period of data collection. In total 1292 Learning Reports were collected.

Analysis The reflections on the learning object were all categorised into the same set of seven categories as in the previous study. Furthermore, all descriptions of the “otherwise, namely…” from the first data collection period were read and analyzed by two researchers. As can be seen in Appendix 2, the frequencies of this category varied between 2.8% and 7.4% (M = 4.6%) for the different variables. Content analysis of the responses to this question uncovered that student teachers used this option when more than one of the multiple choice options was applicable to their learning experience or to describe a more specific example of one of the existing options. Only one-quarter of the descriptions from the alternative option did not resemble one of the existing options or could not be categorised as such. In the multiple correspondence analysis that has been carried out on the categories of the eight variables as described in Study 1, the category “otherwise, namely…” was treated as a missing value, because the content of what students wrote here varied too much to use it as a separate category. The same dimensional structure was found as in Study 1. Dimensions 1 and 2 explained respectively 43.2% and 34.6% of the variances in the eight variables. As can be seen in Figure 4.2, the poles of both dimensions are mirrored compared to the dimensional structure of Study 1. Furthermore, Figure 4.2 shows that even more than with the Open Question Learning Report, the second dimension prospective versus retrospective, particularly separated learning experiences from each other on the active side of dimension 1. 92

Chapter 4

8

RETROSPECTIVE REGULATION

6

4

-2

ACTIVE REGULATION

0

PASSIVE REGULATION

2

PROSPECTIVE REGULATION -2

0

2

4

Figure 4.2. Positioning of all 1292 learning experiences from Study 2 on the two dimensions.

4.4.2 Results study 2 Reliability According to the rule of Cicchetti (1976), for seven categories, 98 ( = 2n2, in which n is the number of categories) observations have to be coded by an independent second researcher to gain a reliable interpretation of Cohen’s Kappa. The inter-rater reliability of the coding of 98 answers to the first question was satisfactory (Cohen’s Kappa = .81). The internal consistency of the scores on the eight variables on the two dimensions has been calculated with Cronbach’s alpha in the same way as for the Open Question Learning Report. Both dimensions showed comparable internal consistency with the open question version: the passive-active dimension .81; the retrospectiveprospective dimension .73.

Validity The same dimension structure was found as with the Open Question Learning Report and the dimensions explained in this data set comparable high amounts 93

of the variances. Although the categories from the qualitative analysis were sometimes positioned slightly differently on the two dimensions than the multiple choice options, the interpretation of the dimensions remained the same. A comparison of Figures 4.1 and 4.2 also showed that, in general, the data are spread in the same way over the two dimensions. The only difference between the spreading of the data of the two studies is that in the second study (Figure 4.2) the variation in scores on the second dimension is smaller on the passive side of the first dimension, than in the first study (Figure 4.1). For validity purposes, an evaluation questionnaire was also distributed (see Section Extra Measurements). The results, obtained from 39 student teachers, showed that they recognised the description of their development as a learner based on the scores on their Learning Reports and ILTP (M = 5.21, SD = 1.08). An indication for predictive validity was calculated by correlating the rubrics-assessment score with the scores on the two dimensions at the last measurement occasion (T3). Six of the nine supervising teacher educators used this system during their final assessment for a total of 58 student teachers. Six of these student teachers were not assessed after one year, because they had not finished all parts of the teacher education programme or they had to do an extra term. Since this was a new assessment system that was still in a pilot phase, we asked the teacher educators to rate the difficulty of the assessment for every student teacher. Their average difficulty rating was 3.46 (SD = 1.42) on a scale from one (very easy) to seven (very difficult). This means that for teacher educators scoring with the rubrics was quite a difficult job. The final assessment scores based on the rating of the teacher educators of the student teachers’ competences varied from 1.83 to 4.00 (M = 2.63, SD = 0.42). The correlations between the aggregated scores over the six learning experiences on both dimensions and the final assessment score have been calculated. Contrary to our expectations, there was no relation found between the score on dimension 1 that describes the passiveness of the regulation and the final assessment score (r = -.06, p = .71). The aggregated score on dimension 2 on T3 showed a moderate correlation of .24, however this was for this small sample only significant at the level of p = .10. This means that a trend was shown that a higher final assessment score was given to student teachers with more retrospectively regulated learning experiences at T3. As an indication for the concurrent validity the aggregated scores on the two dimensions for student teachers were compared with different learning orientations measured by the ILTP. At the three measurement occasions (T1, 94

Chapter 4

T2, and T3), respectively 78, 74, and 69 student teachers completed the questionnaire, from which respectively 74, 71, and 69 student teachers also completed enough Learning Reports to be included in the analysis. The aggregated scores on dimension 1 varied from -1.21 to 1.63 and on dimension 2 from -1.23 to 1,67, with one outlier of 3.59. In Table 4.2, an overview of the means and standard deviation are given for all three measurement occasions separately. As can be seen, there were no large differences on the three occasions in aggregated dimension scores between students with a certain learning orientation. The small differences between the means of the aggregated scores on both dimensions for the four orientations as measured on the three measurement occasions were not significant (all p-values were larger than .05). This means that in this small sample the learning orientation was not related to the activeness or retrospectiveness of the regulation. Table 4.2 Means and Standard Deviations (SD) of the aggregated scores on both dimensions for the different learning orientations on the three measurement occasions Aggregated score on Dim 1 M (SD)

Aggregated score on Dim 2 M (SD)

T1

T2

T3

T1

T2

T3

Open meaningoriented

-0.15 (0.51) N=15

-0.15 (0.70) N=18

0.36 (0.53) N=20

0.10 (0.38) N=15

0.32 (0.67) N=18

-0.00 (0.35) N=20

Closed meaningoriented

-0.06 (0.43) N=21

0.10 (0.66) N=19

0.03 (0.59) N=17

-0.10 (0.54) N=21

0.19 (0.56) N=19

0.23 (1.00) N=17

Reproductionoriented

-0.23 (0.56) N=16

-0.10 (0.60) N=12

-0.05 (0.47) N=18

0.06 (0.62) N=16

0.33 (0.53) N=12

0.20 (0.48) N=18

Survival oriented

0.01 (0.41) N=22

0.13 (0.55) N=22

0.02 (0.48) N=14

0.03 (0.33) N=22

0.23 (0.72) N=22

0.07 (0.51) N=14

Total

-0.10 (0.47) N=74

0.01 (0.63) N=71

0.10 (0.54) N=69

0.01 (0.47) N=74

0.26 (0.63) N=71

0.10 (0.62) N=69

Learning orientation

Discriminative power For the subsequently reported analyzes all learning experiences on the three measurement occasions were analyzed as one sample. All student teachers, 95

except for one, reported learning experiences from different contexts. In total, 330 (25.5%) learning experiences took place at university, 734 (56.8%) learning experiences happened at the practice school, 165 (12.8%) learning experiences occurred in both contexts, and 63 (4.9%) took place in a different context, e.g. at home. From the 1292 learning experiences, 581 (45.0%) experiences were unplanned, 461 (35.7%) were planned and 250 (19.3%) were not planned for the reported experience, but there had been an intention to learn this at some point. On the basis of the scores on both dimensions, all learning experiences could be categorised as showing active retrospective regulation (if dimension 1 < 0.0; and dimension 2 > 0.0), active prospective regulation (if dimension 1 < 0.0; dimension 2 < 0.0) or as being passively regulated (if dimension 1 > 0.0). Since the prospective-retrospective dimension only separated learning experiences from each other on the active side of dimension 1, no distinction was made between prospective and retrospective passive regulation. From all learning experiences, 351 (27.2%) showed the active retrospective type of regulation, 456 (35.3%) showed the active prospective type of regulation, and 485 (37.5%) learning experiences were passively regulated. The combined scores from the six learning experiences on these dimensions together form a regulation pattern. In total seven different regulation patterns could be discerned. These patterns were based on the types of regulation that were present within the six experiences of one student teacher during one data collection period. From the total set of 215 cases (all student teachers during three periods of data collection), in 50 cases the regulation pattern only consisted of one type of regulation. These student teachers were categorised as having an active retrospective regulation pattern (N = 11, 4.1%), an active prospective regulation pattern (N = 18, 6.7%) or passive regulation pattern (N = 21, 7.8%). In 19 (7.0%) cases, the six learning experiences were evenly spread over all three types of regulation and thus the pattern can be labelled a versatile regulation pattern. The other 102 cases showed a pattern with a combination of two regulation types: a pattern of active prospective regulation combined with active retrospective regulation (N = 38, 14.1%); an active retrospective with passive regulation pattern (N = 45, 16.7%), or an active prospective with passive regulation pattern (N = 63, 23.3%).

96

Chapter 4

Usability for teacher education The multiple choice options instead of the open questions should have made the instrument not only less labour intensive for the researcher to process, but also for the student teachers to complete the questions. In the evaluation questionnaire, we asked student teachers to give comments in an open question on their participation in the study. From the 32 student teachers who responded, 22 student teachers spontaneously described that they had liked participating in this study, how much they had learned from participating in this study and reading their personal description of their development as a learner. Six student teachers mentioned something about the time-investment: two of them experienced the time-investment as large, while the other four explicitly mentioned that at first they were afraid it would cost them a lot of time, but that ultimately this was not the case and the personal report made it worthwhile. Seven student teachers highlighted the amount of learning experiences they had to report. Five of them described that they did not always have six learning experiences during the two-week period of data collection, especially towards the end of the programme. On the other hand, one student teacher also mentioned that during the programme the amount of learning experiences did not decrease, but that they were less easy to recognise and thus she had to dig a bit deeper to describe six of them. One student teacher was worried about how her selection of the learning experiences had influenced the results. She reported that she did not learn so much from practice at the end of the programme and therefore reported more learning experiences from university. For the total sample, there were no differences between the three measurement occasions in the distribution of learning experiences over contexts (χ2(6) = 4.24, p = .65).

4.4.3 Conclusion study 2 The underlying dimensional structure that was found with the open version was confirmed with the structured version of the Learning Report. Again the instrument measured to what extent learning experiences were more passively or actively regulated, and whether they were more prospectively or retrospectively regulated. This last dimension turned out to be especially distinctive for the active side of the first dimension. Therefore, three types of regulation could be distinguished: active retrospective regulation, active prospective regulation, and passive regulation. A total of seven regulation patterns were found in the data, slightly more than in the first study. 97

The reliability of the instrument was comparable with the first version and found satisfactory. The extra measurements to obtain indications for predictive and concurrent validity showed no significant relations with learning outcomes and learning orientation. The measures from the evaluation questionnaire showed that the student teachers recognised their outcomes and that a large proportion of the student teachers explicitly mentioned how they liked to participate in the study. For a few student teachers, the time of two weeks was too short to report six learning experiences.

4.5 General discussion The aim of this study was to develop an instrument that could measure student teachers’ regulation activities off-line across multiple and different kinds of learning events from the two contexts of the dual learning programme. Based on the literature, we chose to develop a log with structured questions that could be used as a multiple-event instrument to determine different qualities of student teachers’ regulation of learning by combining the data from different learning experiences. Besides criteria considering the psychometric qualities, aggregation of these multiple events should make it possible to discriminate different qualities of student teachers’ regulation of learning. The Structured Learning Report was based on an open question version that was developed and examined in Study 1. This open question version provided rich descriptions of variation in regulation activities, which were used for the formulation of the multiple choice items of the Structured Learning Report. In case the existing options did not match, it was also possible for student teachers to give their own description. Overall, less than 5% used this option which indicates that the options reflected the variation in regulation activities well. Since most of the student teachers that used the alternative option gave an example of one of the existing options, the formulation of some of the multiple choice options could be reconsidered to further improve the instrument. Although with the Structured Learning Report the rich descriptions of the open question version were lost, the reliability and construct validity did not decrease. The reliability scores of the Structured Learning Report were comparable with the open question version and the same dimensional structure was found. Only the few learning experiences that scored highly in the retrospective passive quadrant in the pre-study were not found again with 98

Chapter 4

the Structured Learning Report. This might be caused by the fact that student teachers with high retrospective passive learning experiences often did not answer questions, which was less likely to happen in the structured version. The Structured Learning Report was suitable for measuring different types of learning experiences from different contexts. Furthermore, the combination of multiple learning experiences into a regulation pattern could discriminate different qualities of student teachers’ regulation of learning at a certain measurement occasion. With the structured version, two more regulation patterns were found than with the open questions. Some of the regulation patterns which were classified as one pattern in the first study could in the larger sample be discriminated as different patterns. Overall, we can conclude that the Structured Learning Report can be considered a valid and reliable instrument to measure and discriminate different qualities of student teachers’ regulation of learning in a dual learning programme. Although the dimensional structure could be interpreted from the literature in combination with the specific characteristics of the learning context, our hypotheses related to the indications for predictive and concurrent validity could not be confirmed. No significant relations were found between the regulation dimension scores and the final assessment score based on the student teachers’ general competence level at the end of the programme and the learning orientations on the three measurement occasions. Future studies are necessary to define the predictive validity of this instrument. To do this, it will be necessary to find a more discriminative measurement of the student teachers’ learning outcomes. The predictive validity of an instrument measuring SRL has been often tested by relating the SRL measurements and learning outcomes derived from the same course or task (e.g., Pintrich & Smith, 1993). For this study only a very general assessment score of the competence level of the student teacher was available. On the other hand, standardised tests, such as a traditional multiple-choice paper and pencil test of teaching have proven to be irrelevant for measuring teachers’ classroom effectiveness (Darling-Hammond & Snyder, 2000). Different forms of multiperspective authentic assessments are promising, but still face many problems (Tillema, 2009). Hopefully, in the future more detailed measurements will become available which could indicate whether the moderate correlation between the retrospectiveness of the regulation and the final assessment score in this data set will be a significant trend in other samples. 99

From previous studies we know that often no relation is found between aptitude and event measurement (Veenman, 2005). Nevertheless, the aptitude and event measurement are both seen as part of the concept of SRL (HowardRose & Winne, 1993; Winne & Perry, 2000). This suggests that if these measurements are part of the same construct, at least some relation should exist. The differences in aggregated scores on the dimensions between the student teachers with a certain learning orientation were in this sample very small and not significant. More research is needed to understand the exact relationship between the learning orientations as measured by the ILTP, and the outcomes of the Structured Learning Report. As Van Hout-Wolters (2000) mentioned, logs and diaries compare students based on different learning experiences. Therefore, the selection of these experiences might be relevant for the outcomes. The number of six learning experiences was chosen to catch variations in the regulation of learning within a student teacher. The answers of the student teachers to the evaluation questionnaire showed that they differed in how easy it was for them to report six learning experiences in two weeks. Further research has to show whether variations in the amount of learning experiences that have to be reported, and the length of the data collection period influence the results. In this study, a multiple event-instrument was developed with the aim of discriminating different qualities of student teachers’ regulation of learning. Results from most single-event instruments describe the quality of regulation during a specific task from which the results can hardly be used to generalise with other situations. Arguments against aptitude instruments have been that with these instruments it is not clear which situations the students have in mind while completing the questionnaire (Perry & Winne, 2006). In the Structured Learning Report, the student teachers did not have to rate themselves as a self-regulated learner or to describe how they acted in general over multiple events. The aggregation over multiple experiences was done by deducing regulation patterns and calculating means based on the results of the multiple correspondence analysis. Therefore, the here reported instrument overcomes the regular problems of aptitude instruments but also has the possibility of discriminating different qualities of regulation on the level of the individual student teacher. The instrument asked student teachers to describe their regulation activities themselves. Critique on self-reports has been that it is not clear whether students actually do what they in a questionnaire say they did or 100

Chapter 4

would do (Veenman, 2005). The study of Winne and Jamieson-Noel (2002) showed that students invariably over and underestimate their use of study tactics. The Learning Report strived to overcome these problems with the following characteristics. First, since the student teachers could access the instrument at any time, the learning experiences could be measured as closely as possible to the actual learning events (Van Hout Wolters, 2006). Second, student teachers first had to describe the learning experience in their own words, which made it difficult for student teachers to subsequently pick a different activity from the multiple choice options than actually fitted their experience. Furthermore, the instrument focused on identifying the qualitative different regulation activities that student teachers used for every learning experience and thus did not ask them to estimate frequencies of use. Therefore, although self-reports were used, in our opinion this did not affect the reliability and validity of the results to a great extent. In conclusion, an instrument has been developed that can be used in teacher education programmes to determine the quality of student teachers’ regulation of learning. Since the Structured Learning Report had comparable psychometric qualities, it outperformed the more labour-intensive open question version. The two-step procedure that has been used to develop this type of instrument could be exemplary for developing instruments for measuring students’ SRL in other dual learning programmes than teacher education, or for other aspects of learning. This study showed how a multipleevent instrument can be used to measure regulation of learning in multiple contexts for various learning experiences at the same time, without the necessity of relying on students’ ability to rate themselves across all these different experiences. In this way, this instrument tried to overcome the limitations of traditional aptitude and event measurements and in this way attempted to bridge the gap between these two measurement traditions.

101

Appendix A: Questions of the Open Question Learning Report 1. What did you learn? 2. Did you plan to learn this, and if so, why did you want to learn this? 3. Did you expect to succeed in learning this and what made you think you would (not) succeed in learning this? 4. How did you learn this? 5. Why did you learn it in this way? 6. How did you realise that you had learned something? 7. If you look back, about which aspects are you satisfied and what would you do differently next time? 8. How will you proceed with this learning experience?

102

Chapter 4

Appendix B: Questions and multiple choice items of the Structured Learning Report including frequencies (freq). 1. What did you learn? Open question, categorised in terms of the following reflections on the learning content: Multiple choice options

Freq

%

A

Reflection on learning content in terms of a rule of thumb.

220

17.0

B

Reflection on learning content in terms of factual knowledge.

302

23.4

C

Reflection on learning content in terms of procedural knowledge.

78

6.0

D

Reflection on learning content in terms of own learning or identity as a teacher.

277

21.4

E

Reflection on learning content in terms of a specific teaching practice.

139

10.8

F

Reflection on learning content in terms of theory of practice.

194

15.0

G

No reflection in terms of learning, only description of an experience.

82

6.3

1292

100.0

Total

2a. Did you plan to learn this? Multiple choice options

Freq

%

A

No, I did not plan to learn this (proceed with question 3).

581

45.0

B

Not specifically for this moment, but I had an intention to learn this.

250

19.3

C

Yes

461

35.7

Total

1292

100

Multiple choice options

Freq

%

A

I was unsatisfied about a previous experience.

206

15.9

B

I was curious about something.

95

7.4

C

Others stimulated me to develop myself in this.

125

9.7

D

I wanted to prepare myself for future possible experiences.

92

7.1

E

I wanted to practice with something.

101

7.8

F

Otherwise, namely…

95

7.4

(skipped, because of an unplanned learning experience)

578

44.7

1292

100.0

2b. What was the main reason to learn this?

103

3. There are different ways to learn things. Not all ways are always applicable to every situation. Please, choose the description that fits your experience best. I learned something by…. Multiple choice options

Freq

%

A

… I don’t know actually.

12

0.9

B

… doing it or experiencing it.

352

27.2

C

… experimenting something.

174

13.5

D

… evaluating what went well and wrong in my lesson or another situation.

93

7.2

E

… analyzing my and others’ role in a situation.

85

6.6

F

… getting information.

269

20.8

G

… getting feedback from others.

186

14.4

H

… observing how others do something.

46

3.6

I

Otherwise, namely…

75

5.8

1292

100.0

Total

4a. Did you choose beforehand this way of learning? (In the questionnaire, this question is only asked to people who reported a planned learning experience): Multiple choice options

Freq

%

A

No, this was no conscious choice (proceed with question 5).

307

23.8

B

Yes, I thought about that beforehand.

403

31.2

(skipped, because of an unplanned learning experience)

582

45.0

Total

1292

100.0

4b. You just noticed that you chose your way of learning beforehand. Why did you choose THIS way of learning? Multiple choice options

Freq

%

A

I don’t know.

19

1.5

B

It is not possible to learn it in another way.

125

9.7

C

Someone else suggested to me to learn it this way.

74

5.7

D

This was the easiest or the fastest way to learn it.

75

5.8

E

Compared with other ways of learning, this way of learning often works well for me.

45

3.5

F

Otherwise, namely …

65

5.0

(skipped, because it was no conscious choice)

307

23.8

(skipped, because of an unplanned learning experience)

582

45.0

Total

1292

100.0

104

Chapter 4

5a. Did you expect to succeed in learning this? (In the questionnaire, this question is only asked to people who reported a planned learning experience) Multiple choice options

Freq

%

A

Yes

220

17.0

B

No

11

0.9

C

I didn’t know, but I hoped to succeed (proceed with question 6).

178

13.8

D

I didn’t think about that beforehand (proceed with question 6).

55

4.3

(skipped, because of an unplanned or unintentional learning experience)

828

64.1

Total

1292

100.0

5b. Why did you expect (not) to succeed in this? (In the questionnaire, this question is split up in a positive and negative version): Multiple choice options

Freq

%

A

I was (not) confident in myself to succeed.

123

9.5

B

I was well prepared.

42

3.3

C

The last time I learned something in this WAY, it also worked out well/did not work out well.

23

1.8

D

The last time I learned something in this CONTEXT, it also worked out well/did not work out well.

5

0.4

E

Otherwise, namely…

38

2.9

(skipped, because did not think about it)

233

18.1

(skipped, because of an unplanned or unintentional learning experience)

828

64.1

Total

1292

100.0

6. At what moment did you realise that you had learned something? Multiple choice options

Freq

%

A

I don’t know.

20

1.5

B

The moment I experienced that it worked out well.

266

20.6

C

The moment I experienced that it did NOT work out well.

68

5.3

D

The moment I saw or heard the reaction of others.

107

8.3

E

The moment I received feedback.

147

11.4

F

The moment I reflected on my experience.

165

12.8

G

The moment I realised that I received new information.

232

18.0

H

The moment I became aware of my own behaviour.

97

7.5

I

Otherwise, namely …

36

2.8

Missing values (due to a mistake in a skip logic)

154

11.9

Total

1292

100.0 105

7a. When you look back on this learning experience, is there something you are unsatisfied about? Multiple choice options

Freq

%

A

No (proceed with question 8)

1029

79.6

B

Yes

263

20.4

Total

1292

100.0

Freq

%

7b. What are you especially unsatisfied about? Retrospectively,…. Multiple choice options A

… I would have wanted to learn this earlier in my development.

86

6.7

B

… I would have wanted to prepare myself better.

31

2.4

C

… I would have wanted to tackle things differently during this experience.

49

3.8

D

… I would have liked to learn this in a different way.

3

.2

E

… I would have wanted my students to behave differently.

25

1.9

F

… I would have hoped that others would cooperate better.

15

1.2

G

Otherwise, namely…

54

4.2

(skipped, because totally satisfied)

1029

79.6

Total

1292

100.0

Multiple choice options

Freq

%

A

I have no new plans (yet).

121

9.4

B

It did not work out the way I wanted, so I am going to try again.

27

2.1

C

I have exactly figured out what I will do next time in a comparable situation.

76

5.9

D

I want to consolidate what I have learned.

160

12.4

E

I want to improve further what I have learned.

308

23.8

F

I want to apply in practice what I have learned.

367

28.4

G

I want to try out what I have learned in a different situation.

93

7.2

H

Based on what I have learned, I have formulated a new learning goal for myself.

85

6.6

I

Otherwise, namely…

55

4.3

1292

100.0

8. How do you proceed with this learning experience?

Total

106

instructed

individual

completed

meaning-oriented

longitudinal

focused

regular

indication

willms

become away

slope vermunt

months

lodewijks

versus

types

programme

semester

experiences

context amount responsibilities

increase

information

abbreviations

growth

choice

education

decrease

random

preferences

performance description

internship

terms

external

tillema

university

analysis specific distribution conceptions

prepared

accepted

displayed

monitoring

conceptual

environment

patterns

across

became

results

chapter

evaluation

prog

values

learner

differences regarding

defined

van petegem

linear

preoccupied

willett

rate

oosterheert

learning report

stable

student teachers' teaching pre-service variance predictor combined started occasions hagger variables following quadratic

personal

netherlands

influence problems

students

development sources

correlations

standard assumed deviation

scales

independent frame

prospective lessons

model

skills

questionnaire

multilevel average

research

theory

iltp

significantly

characterised

srl

active

important

testing

outcomes quality

self-regulated academic

discussion previous

conclude per

reflecting scores

opportunities

measurement

directing

collected

intraclass

related included

higher positively

expected

secondary

strategy goals

learning paid

passive

regulation

interns

planning

theoretical

explored

knowledge

reference

improvement

lack

lifelong

intern-track

extent

compared report

situations

ideas

aspects

participants

described

reproduction-oriented

teacher

multiple

data

thinking

dual

fitted

vermetten

intercept institutes

categorised

retrospective

phases

predicted

significant

means nevertheless job-track explain giving dimensions role

orientations explanation

tracks

dependent

structured

5 Student teachers’ development in self-regulated learning throughout a dual learning programme5 Abstract One of the tasks of pre-service teacher education is to support student teachers in developing the conceptions and skills necessary for lifelong learning. It has been assumed that teacher education programmes can play a major role in developing SRL in student teachers by giving opportunities for directing one’s own learning. This study aimed at describing student teachers’ development in self-regulated learning throughout a dual learning programme. A longitudinal design with three measurement occasions was adopted in this study, in which 81 (90%) student teachers from two tracks of one dual teacher education programme participated. Their learning orientations were measured with a questionnaire and their regulation activities by means of multiple structured question logs. Longitudinal multilevel analyses showed that student teachers became more passive in their regulation throughout the programme. Only one third of the student teachers changed in the direction of independent meaning-oriented learning, while the others changed away from it or kept the same orientation over time. Also differences between the job-track and interntrack of the programme in learning orientations were found. The results show that opportunities for directing one’s own learning might be a necessary, but not a sufficient condition to increase student teachers’ conceptions and skills to become a self-regulated lifelong learner.

This chapter has been submitted in adapted form as: Endedijk, M., Vermunt, J.D., Meijer, P.C., & Brekelmans, M. Student teachers’ development in self-regulated learning throughout a dual learning programme. 5

109

5.1 Introduction Self-regulated learning is a means as well as an end in pre-service teacher education. One of the central tasks of pre-service preparation is to support and challenge student teachers to develop the conceptions and skills necessary for active lifelong learning (Feiman-Nemser, 2001; Hagger, Burn, Mutton, & Brindley, 2008; Kremer-Hayon & Tillema, 1999). When learning is no longer supported and fed by teacher education programmes, self-regulated learning is an important prerequisite to be able to reflect on and innovate teaching habits in the dynamic environment of a classroom (Hammerness et al., 2005; Niemi, 2002). It has been assumed that teacher education programmes can play a major role in developing SRL in student teachers by giving opportunities for directing one’s own learning (Kremer-Hayon & Tillema, 1999). Consequently, contemporary teacher education programmes are designed in a way that already draws upon a high degree of self-regulated learning. Especially in dual learning programmes, in which student teachers learn in parallel at university and in practice, they need the capacity to actively integrate theory in their practice, to identify their strengths and weaknesses in order to formulate their own learning needs and choose suitable learning strategies. In this way, student teachers’ active learning skills determine to a great extent the learning potential of these programmes (Mutton, Burn, & Hagger, 2010). Although learning orientations and learning activities are amenable to change under influence of the context (Tynjälä, 1997; Vermetten, Lodewijks, & Vermunt, 1999), the question is whether a programme that requires selfregulated learning indeed leads to the development of these learning skills in student teachers. Previous studies have shown that student teachers differ in their learning in such dual programmes. These studies have revealed that only a minority of student teachers have an active, independent and meaningoriented way of learning (Hagger et al., 2008; Oosterheert & Vermunt, 2001). Nevertheless, longitudinal studies into how student teachers develop in their self-regulated learning are scarce. Therefore, this study aims at describing the development in student teachers’ self-regulated learning throughout a dual learning programme.

5.2 Theoretical framework The definition of self-regulated learning (SRL) has been shifting during the past decades (Alexander, 2008), but the origins of the work of Flavell (1976) 110

Chapter 5

and Brown (1978) can still be recognised in the current conceptualisations. The contemporary broad definition of the concept of self-regulated learning includes metacognitive knowledge as well as regulation of learning (Dinsmore, Alexander, & Loughlin, 2008). The metacognitive knowledge part of SRL has been regarded as the more stable part, whereas regulation of learning is viewed as the more dynamic part of SRL that responds to the demands of the learning context (Brown, 1987; Pintrich, 2004). Historically, metacognitive knowledge was focused on specific factual knowledge students had about their learning. Nowadays, this more stable part of self-regulated learning is also studied as the conceptions students have about learning in general, their learning and thinking activities, oneself as a learner, and their regulation conceptions (Vermunt, 1996). For the context of pre-service teacher education, Oosterheert and Vermunt (2001) used the concept of learning orientations to denote typical combinations of learning conceptions and preferences for learning- and regulation activities. If learning orientations are measured without a specific learning situation in mind, they refer to the learner’s general disposition to learning (Vermetten, Lodewijks et al., 1999). Below, findings of previous studies into individual differences and development of both parts of SRL, conceptualised as learning orientations and regulation of learning will be discussed in the context of student teacher learning.

5.2.1 Student teachers’ learning orientations Individual differences in learning orientations Research on knowledge and beliefs about learning and cognition started with the work of Säljö (1979) describing students’ differences in how they see the fundamentals of learning. This work gave rise to a large field of research into students’ learning conceptions, learning approaches, learning styles, and learning orientations, all covering different aspects of the process of studying (Entwistle & McCune, 2004). The study of Oosterheert and Vermunt (2001) has shown that the following dimensions are relevant for describing individual differences in orientations towards learning to teach: the dimension meaningoriented versus performance-oriented learning showed differences between student teachers in whether they learn by questioning and changing their frame of reference, or whether they are directed at improving performance through gathering practical suggestions. The dimension open versus closed learning showed differences in how student teachers approached their problems: whether they acknowledged their problems and tried to find 111

solutions independently or that their problems remained more implicit and they needed others to solve those (Oosterheert & Vermunt, 2001). Student teachers could be categorised in the following four learning orientations (Oosterheert, Vermunt, & Veenstra, 2002): 1. Inactive or survival orientation. These student teachers stress that all you need for learning to teach is a lot of teaching practice and experience. They do not appreciate the help of others to become aware of their teaching, but also do not think that they are able to regulate their learning themselves. They hardly use the available sources in their learning environment and are very avoidant and not preoccupied with bad lesson experiences. 2. Closed (or dependent) reproduction orientation. These student teachers are focused on improving their teaching performance within their actual frame of reference. They are not directed at further developing this frame of reference which results in limited use of available sources. They acknowledge bad lesson experiences and have serious worries about these. 3. Closed (or dependent) meaning orientation. These student teachers try to extend their frame of reference and depend on external sources in doing so, which they highly value. They do not trust their own perceptions and thinking; others have to help them to interpret their experiences. They are extremely preoccupied with their bad teaching experiences. 4. Open (or independent) meaning orientation. These student teachers are most independent in learning to teach: they try to develop their frame of reference, make broad use of external sources and are highly selfregulative. They define problems of learning to teach not only as problems of performance, but also of meaning. They are not very preoccupied with bad lesson experiences (Oosterheert, Vermunt, & Veenstra, 2002). Not only for academic learning, but also for learning to teach, an independent meaning-oriented learning orientation is regarded as essential in being prepared for life-long learning (Bakkenes, Vermunt, & Wubbels, in press; Oosterheert, 2001).

Development in student teachers’ learning orientations Oosterheert et al. (2002) expected learning orientations not to change spontaneously, but that with the right interventions the quality of student teachers’ learning processes can be improved. Oosterheert (2001) hypothesised that if student teachers change their learning orientation, this will be probably occur from survival or reproduction-oriented towards dependent meaning112

Chapter 5

oriented and ending in independent meaning-oriented learning. In a longitudinal study, Donche and Van Petegem (2009) have studied changes in pre-service student teachers’ learning orientations towards academic learning, measured by the Inventory of Learning Styles (Vermunt, 1998). Their study showed indeed an increase in meaning-oriented learning in the academic part of the teacher education programme between the first and third year of study (Donche & Van Petegem, 2009). These results are also in line with most of the longitudinal studies done with the ILS in other parts of higher education: advanced students show more deep-level processing, more self-regulated learning and higher levels of critical thinking in comparison with novices, although at the same time the level of reproduction-oriented learning remains the same (Severiens, Ten Dam, & Van Hout Wolters, 2001; Vermetten, Vermunt, & Lodewijks, 1999; Vermunt & Vermetten, 2004). On the other hand, in a small-scale study, Hagger, Burn, and Mutton (2008) also described student teachers who became more focused on their own practice as the only source for learning and who became less ambitious as learners, especially in the final part of the programme. This suggests that some student teachers might as well develop away from independent meaning-oriented learning.

5.2.2 Student teachers’ regulation of learning Individual differences in regulation of learning From research into students’ regulation of academic learning we know that students differ in their quality of regulation. The study of Schunk and Zimmerman (1998) showed that compared to naïve self-regulators, skilful selfregulators set higher quality goals, have higher self-efficacy, can instruct themselves and monitor their learning process instead of using selfhandicapping strategies, seek more opportunities for self-evaluation and reflect systematically on their learning experiences which positively influences new forethought processes (Schunk & Zimmerman, 1998). Both teachers and students draw on SRL strategies to cope with situational demands in the classroom (Randi, 2004). Research into SRL has, however, focused primarily on student learning in traditional instructional settings. One of our previous studies focused therefore on identifying the nature and individual differences of student teachers’ regulation of learning. This study showed that differences in the nature of regulation of learning experiences in dual learning programmes could be best described according to the following two dimensions (see also Chapter 2): 113

1. Active versus passive regulation. An actively regulated learning experience showed more deliberate choices of student teachers in steering their own learning. For example, in such learning experiences student teachers made their own decisions for a learning strategy, but also actively used information from others in their reflections on the learning content, the learning process and their own role in it. Learning experiences that were passively regulated by student teachers were characterised by lack of regulation or reliance on external regulation in all phases of the learning process. 2. Prospective versus retrospective regulation. In a prospectively regulated learning experience, the focus of the regulation activities was on the forethought phase of the learning process. These learning experiences were characterised by high quality of planning, goal setting and argumentations for choosing a learning strategy were given. The regulation of monitoring, reflection, and evaluation was more superficial and more focused on the learning outcomes than on the learning process. The retrospectively regulated learning experiences were often unplanned, so no goal-setting or deliberate thinking about learning strategy and self-efficacy could have taken place. Nevertheless, retrospective regulation was characterised by more deliberate deep ways of monitoring, evaluation and reflection on their learning process and outcomes. By studying the regulation of multiple learning experiences per student teacher, different patterns were distinguished describing individual differences in the amount of the activeness (or passiveness) and prospectiveness (or retrospectiveness) of their regulation of learning (see Chapter 3).

Development in student teachers’ regulation of learning Individual changes in regulation activities as performed in concrete learning experiences are seldom studied. This might stem from the view that regulation of learning is not a personal trait, but that it evolves in the interaction with the context and the specific learning task (Zimmerman, 2002). In contrast, several studies have shown how self-regulated learning of students in primary and secondary school can be developed with specific interventions (Dignath & Büttner, 2008). To conclude, under the right circumstances students in general can become more active in their regulation, however, information about individual development in the use of regulation activities in non-experimental settings is still lacking. 114

Chapter 5

5.2.3 Influence of the teacher education programme on student teachers’ self-regulated learning In response to the criticism of being overly theoretical, the design of preservice teacher education programmes has changed during the past decades from mainly learning in higher education institutes to more school-based teacher education (Maandag, Deinum, Hofman, & Buitink, 2007). In contrast to the UK, the move towards school-based teacher education in The Netherlands has developed more recently and is less far-reaching (Maandag et al., 2007; Ten Dam & Blom, 2006). Most of the post-graduate teacher education for upper secondary education in The Netherlands can nowadays be best described as a dual learning programme, in which student teachers learn in parallel in practice and at the university. This programme includes models in which student teachers are employed as a teacher and have all the responsibilities of a regular teacher (Buitink, 2009), as well as models where student teachers do an internship and are more gradually exposed to the teaching profession. Several researchers and teacher educators have expressed their worries about programmes in which a large amount of time is spent on teaching practice, since this reflects a view on learning that teaching can be learned by doing and imitation without making explicit the underlying principles (e.g., Buitink, 2009; Hagger et al., 2008). The learning potential of the practice part of the programme is considered to be dependent on the space and time student teachers have for interpreting their experiences, to test theories and to investigate and analyse their teaching problems while critically examining their different conceptions (Feiman-Nemser, 2001; Hagger et al., 2008). Observation, apprenticeship and guided practice are important pedagogies to achieve that aim (Feiman-Nemser, 2001). In particular, with programmes in which student teachers have all the responsibilities of a regular teacher, there might not be enough time for developing lifelong learning skills. To conclude, the learning context of having a job or being an intern might influence the development of student teachers’ self-regulated learning.

5.2.4 The present study In this study, a longitudinal design is adopted to study student teachers’ development in the two parts of self-regulated learning (learning orientations and regulation of learning). We view and assess learning orientations as student teachers’ combination of learning conception and preferences for certain learning and regulation activities (Oosterheert, Vermunt, & Denessen, 115

2002). Student teachers’ regulation of learning is defined as the regulation activities they perform in multiple concrete learning experiences. Furthermore, we will investigate to what extent the type of programme is related to the development of student teachers’ learning orientations and regulation activities. The following research questions have been formulated: 1 How do student teachers’ regulation of learning and learning orientations change during a one-year dual teacher education programme? 2 To what extent can the programme track (job-track or intern-track) explain differences in student teachers’ regulation of learning and learning orientations and in how these change throughout the programme? Drawing upon the literature as described in the Theoretical Framework, we have the following expectations related to these questions: 1a Student teachers are expected to change during a dual learning programme towards more active regulation of learning (e.g., Kremer-Hayon & Tillema, 1999). 1b Student teachers are expected to change during a dual learning programme towards more independent meaning-oriented learning (Oosterheert, 2001). 2 Student teachers following an intern-track of the teacher education programme are expected to develop a more active way of regulation and more meaning-oriented way of learning compared to the student teachers having a paid job as a teacher (e.g., Hagger et al., 2008).

5.3 Method 5.3.1 Design Traditionally, individual change has often been measured as the difference between “begin” and “end”, measured with a two-wave research design (Willett, 1997), which also has been applied in research on learning orientations (e.g., Donche & Van Petegem, 2009; Vermetten, Vermunt, & Lodewijks., 1999). Such designs have been important for detecting development in group means over time, but are not very suitable for accurately modelling individual differences in growth (Marsh, 2007; Raudenbush & Bryk, 2002; Singer & Willett, 2003). Hierarchical linear modelling or in other words, multilevel analysis, has developed into a more powerful technique for studying individual growth (Raudenbush & Bryk, 2002), since it makes it possible to accommodate any number of measurement occasions and to include also individuals with incomplete data (Hedeker, 2004; 116

Chapter 5

Hox, 2000; Willett, 1997). The choice for the number of measurement occasions depends on the expected memory effect of the instruments and on the expected rapidness of change; however it should include at least three occasions (Boyle & Willms, 2008; Hox, 2002; Singer & Willett, 2003). Since learning orientations are expected to be quite stable, though changeable (Oosterheert, 2001; Vermunt, 2005), this study includes three measurement occasions. Data regarding regulation of learning were collected about six learning experiences per student teacher per measurement occasion, to capture the variation in regulation due to contextual influences. As can be seen in Figure 5.1, the first measurement occasion took place after about three months of the start of the programme. This amount of time was considered to be the minimum for the student teachers to form their orientations for learning to teach in the new dual learning environment. The other two measurement occasions took place with intervals of three months. aug

sept

oct

Start

nov

dec

T1 Measurement occasion

jan

feb

mar

apr

T2

may

T3

june

july

aug

End

School change for interns

Holiday

Figure 5.1. Overview of the longitudinal research design. The figure shows when the measurement occasions T1, T2, and T3 took place related to the holiday periods and the school change of the interns in January.

5.3.2 Context The development of student teachers’ SRL has been studied in one postgraduate one-year teacher education programme in The Netherlands preparing for subject-teaching in upper secondary school. Two different tracks of this programme were included and compared in this study: the job-track and the intern-track. The content of the academic part of the programme is the same for both tracks. All student teachers attend in the first semester weekly general and subject-specific classes at the university organised by teacher educators. In these classes, they also meet other student teachers to discuss and learn about several aspects of the teaching profession. In the second semester, these meetings have a lower frequency. All student teachers have much 117

freedom in designing their personal curriculum, based on their prior experiences and concerns. For example, they can decide for themselves about which topics they want to attend extra theoretical lectures. They keep a portfolio in which they include professional products, make self-evaluations and a personal development plan to direct their learning processes. In this personal development plan they also have to reflect on how they take responsibility for their own development. Their assessment at the end of the programme is based on their portfolio and a final interview. Regarding the practice part, student teachers doing an internship are gradually exposed to the job. In the first semester they do their internship in a triad. They observe experienced teachers and their peer student teachers while having their first teaching experiences. In the last months of this first semester, they teach about five lessons a week. In the second semester, the interns change schools and have their second internship with more responsibilities as a teacher, including an increase of lessons they teach. The student teachers with a job have all the responsibilities of a regular teacher right from the beginning and often teach more than ten lessons a week. Student teachers from both tracks are supervised at their practice school by an experienced teacher.

5.3.3 Participants All 90 full-time student teachers from the two tracks were asked to volunteer in this study. All student teachers were initially willing to participate; however, five student teachers cancelled their participation, because of lack of time, illness, pregnancy, other expectations, and one left the teacher education programme. Four other student teachers did not cancel their participation explicitly, but also did not fill out a single questionnaire. The final set of 81 participants (response rate 90%) consisted of 24 (29.6%) male and 57 (70.4%) female student teachers from eighteen different general secondary school subjects. From these student teachers, 47 (58.0%) started the programme as an intern, while 34 (42.0%) student teachers followed the programme as a paid teacher in the job-track. Student teachers having a job were significantly older (M = 28.06, SD = 4.74) than the interns (M = 24.21, SD = 2.19, t(43.20) = -4.40, p < .001) and 88.9% of the interns had no prior experience as a teacher in secondary education, while this was the case for only 48.5% of the student teachers following the job-track (χ2(1) = 15.35, p < .001). On the other hand, 46.7 % of the interns had followed a short introduction course for teacher education (7.5 ECTS) in which they taught a few lessons in a school for secondary 118

Chapter 5

education. From the student teachers with a job, only 27.3% followed this introduction course (χ2(1) = 3.03, p = .08). In the first semester, the intern-track student teachers gave significantly fewer lessons than the job-track student teachers (U = 25.50, z = -7.44, p < .001). In the second semester, the differences between the two tracks in the number of lessons was smaller, but still significant (U = 240.50, z = -4.22, p < .001). During the first semester, the jobtrack student teachers spent also more time per week on the practice part than the interns (U = 403.50, z = -3.50, p < .001). In the second semester, this difference was not significant anymore (U = 480.50, z = -1.17, p = .24). The time spent on the university part of the educational programme was similar for the two tracks during both semesters.

5.3.4 Instrumentation Measuring regulation of learning For the purpose of this study an instrument was developed to measure student teachers’ regulation of learning, called the Structured Learning Report. The Structured Learning Report is a multiple-event measurement, meaning that regulation of learning is measured from student teachers’ reports of different concrete learning experiences (n = 6). The instrument consisted of seven multiple choice questions reflecting different aspects of the process of selfregulated learning: student teachers’ goal orientation, notions of self-efficacy, learning strategy choice, forethought, control of the learning strategy, monitoring of the learning results, and evaluation of the learning experience (see also Chapter 4, Appendix B). The choice and formulation of the items was based on a previous open question version, as described in Chapter 4. The Structured Learning Report started with one open question: What did you learn? In this way, the student teachers were able to describe in their own words their reflections on their learning outcomes. The answers on this open question were all categorised by the researcher into seven qualitative different categories of reflections on learning outcomes. Inter-rater reliability procedure was carried out by a second independent researcher, resulting in a Cohen’s Kappa of .81. The total set of eight variables reflected the four phases of selfregulation as described by Pintrich (2000). With help from multiple correspondence analysis (Mair & De Leeuw, 2008) the categorical scores for every learning experience on the eight variables were transformed to scores on underlying dimensions in the data. The underlying structure in the data set turned out to be equal to the two dimensions of our previous study: all 119

learning experiences could be described in terms of their activeness or passiveness of regulation and in terms of whether the regulation was more prospective or retrospective (see also the description in the Theoretical Framework). The Eigenvalues of Dimensions 1 and 2 were respectively 3.45 and 2.77, explaining 43% and 35% of the variance. As can be seen in Table 5.1, the reliability of these dimensions was satisfactory. For a more detailed description of the development and examination of the Structured Learning Report, we refer to Chapter 4. The aggregated scores on both dimensions were used as a measurement of student teachers’ quality of regulation of learning across the six learning experiences on each measurement occasion.

Measuring learning orientations The student teachers’ learning orientations were studied with the revised version of the Inventory Learning to Teach Process (ILTP), developed by Oosterheert et al. (2002). This questionnaire had been used and validated in The Netherlands with 382 student teachers (Oosterheert, 2001) and in Belgium with 366 student teachers (Donche & Van Petegem, 2005). The instrument consists of ten scales with Likert-scale items reflecting their learning conception, preferences for learning and regulation activities, and emotion regulation. The different scales, including reliability scores and some descriptives are mentioned in Table 5.1. A 7-point Likert scale was used instead of the original 5-point scale, in order to increase the sensitivity for changes throughout the programme. For a more complete description of the development and description of the instrument and scales we refer to Oosterheert et al. (2002).

120

Chapter 5

Table 5.1 Overview of variables across all three measurement occasions (Cronbach’s K, means and standard deviations, and intraclass correlations (ICC)) Variables (abbreviation)

Items

Cronbach’s α

M (SD)

ICC

Structured Learning Report - Regulation of learning (aggregated dimension scores1) Passiveness of regulation (REG-PS)

8 var.

.83

0.00 (0.55)

.32

Retrospectiveness of regulation (REG-RE)

8 var.

.72

0.12 (0.58)

.15

Practising and testing (LC-PT)

9

.70

4.95 (0.61)

.67

Strong self-determination in performance improvement (LC-SD)

3

.54

3.89 (0.82)

.49

Raising consciousness under external control (LC-CE)

7

.73

5.41 (0.73)

.65

ILTP - Learning conception (min.=1, max.=7)

ILTP - Preference for learning and regulation activities (min.=1, max.=7) Proactive, broad use of the mentor (LA-PM)

6

.86

4.93 (1.09)

.62

Independent search for conceptual information (LA-IC)

5

.86

3.48 (1.24)

.68

Actively relating theory and practice (LA-TP)

5

.89

4.50 (1.18)

.80

Developing views/ideas through discussion (LA-VD)

5

.82

4.40 (1.14)

.68

Pupil-oriented evaluation criteria (LA-PE)

3

.63

5.65 (0.67)

.42

Avoidance (ER-AV)

5

.72

1.88 (0.61)

.50

Preoccupation (ER-PO)

4

.82

4.52 (1.42)

.60

ILTP - Emotion Regulation (min.=1, max.=7)

1 These scores are the aggregated values of the dimension scores of six learning experiences. These scores ranged for REG-PS from -1.21 till 1.63 and for REG-RE from -1.23 to 1.67, with one outlier of 3.59.

5.3.5 Procedure The data collection took place in three periods of two weeks in which student teachers were instructed to report online six different learning experiences during every period in the Structured Learning Report and to fill out the ILTP. After these two weeks reminders were sent to collect the missing reports and questionnaires. They were instructed to describe different kinds of learning experiences as part of their development as a teacher in the six Learning Reports: two learning experiences that had taken place (at least partially) in the context of the teacher education institute, and at least two that had taken place (at least partially) in the context of the practice school, and two free of choice. Furthermore, they were told that they could report planned as well as 121

spontaneous learning experiences, and both successful and unsuccessful learning experiences. As a reward for their participation, they received a personal description of their development as a learner at the end of the study. On the three measurement occasions, respectively 78, 74, and 68 student teachers completed the ILTP. In total, 66 student teachers completed all 18 Learning Reports, five student teachers participated only in two, and six student teachers only during one measurement occasion. Four student teachers completed fewer than four Learning Reports, which was too few to be included in the analysis. For the Learning Reports, this resulted in 75 participants at T1, 71 participants at T2, and 69 participants at T3. In this way 1292 Learning Reports were collected.

5.3.6 Analysis We used the following combination of analysis techniques to study student teachers’ development in SRL in the two tracks of the dual learning programme: First, (univariate) longitudinal multilevel analysis was used to describe the individual change in scores on the twelve dependent variables and effects of the programme tracks on this change. These dependent variables included ten scales of the ILTP and the two aggregated dimension scores reflecting the passiveness (dimension 1) and the retrospectiveness (dimension 2) of the regulation of learning. These variables, including the abbreviations which will be used in the Result Section, are shown in Table 5.1. The programme track that student teachers followed (PROG), varying between the intern-track (PROG = 1), and the job-track (PROG = 0) was added as a predictor variable. Second, to obtain an integrated view of the change in scores on the total set of twelve variables, a multivariate multilevel model is needed. However, estimating a longitudinal multivariate multilevel model with twelve variables needs more measurement occasions and a sample size much bigger than available (Muthén & Muthén, 2002). A more comprehensive view on the change in the two parts of self-regulated learning was obtained by determining change within individuals on the four learning orientations and regulation patterns. Below, both methods of analysis will be described in more detail.

Longitudinal multilevel analyses To study student teachers’ change trajectories we considered a two-level model with measurement occasions (level 1) nested within student teachers (level 2) 122

Chapter 5

(Snijders, 1999). Different growth models were tested to choose the simplest growth model that still fitted the data well. Our study, with three measurement occasions, allowed us to compare the fit of three types of growth models: a model with no change over time, a linear growth model, and a quadratic growth model (Boyle & Willms, 2008). Each model is characterised by an intercept, which represents a student teacher’s baseline, and a slope, which represents a student teacher’s growth rate (Boyle & Willms, 2008; Willett, 1997). In every model intercepts and slopes can be assumed fixed (the same for all students) or random (differences between students) (Hedeker, 2004). To answer the first research questions the following models with increased complexity were tested in the following order: 1. No change over time model - random intercept; 2. Linear growth model - random intercept, fixed slope; 3. Quadratic growth model - random intercept, fixed slope; 4. Linear growth model - random intercept, random slope; 5. Quadratic growth model - random intercept, random slope. After the growth model had been selected to describe the within-individual change (research question 1), the time-invariant predictor PROG (0 = job, 1 = intern) was added, to explain the individual differences in their intercepts and slopes (research question 2). A likelihood ratio test (Hedeker, 2004) and also the statistical significance of the parameters were used to determine whether a more complex model had to be accepted (Boyle & Willms, 2008). The time variable was defined in a way that T1 had value 0.

Analyses of learning orientations and regulation patterns The learning orientations and regulation of learning can both also be expressed in a pattern of the scores on the different scales or dimensions. With the help of z-scores of the ten scales of the ILTP, it could be calculated which learning orientation as defined by Oosterheert (2001) was the closest to the score pattern of the student teacher. In this way, for every student teacher on every measurement occasion his or her learning orientation was defined: survivaloriented, reproduction-oriented, dependent meaning-oriented or independent meaning-oriented learning. To give an indication: independent meaningoriented learning is related to high scores on the scales strong self-determination in performance improvement (LC-SD), independent search for conceptual information (LA-IC), actively relating theory and practice (LA-TP), and developing views/ideas through discussion (LA-VD) and a low score on the emotion-regulation scale preoccupation (ER-PO) (see also Oosterheert et al., 2002). 123

To answer the first research question, the directions of changes in student teachers’ learning orientation over time was described. This change was categorised as positive, when change was found towards independent meaning-oriented learning and as negative when the change was away from this learning orientation, according to the development hypothesis of Oosterheert (2001). Chi-square tests were used to describe differences between the two tracks in the categorisation of their development in learning orientation. The positions of the six learning experiences on the two dimensions formed the regulation pattern of a student teacher on each measurement occasion. In total seven different regulation patterns could be discerned (see also Table 5.4). The following heuristics were followed for categorisation: First all six learning experiences were classified in one of the following regulation types: passive regulation, prospective active regulation, or retrospective active regulation. These three types of regulation were marked off by the axes of the dimensions. The regulation types that a student teacher reported in at least two learning experiences determined his or her regulation pattern. In case the learning experiences could be typified by only one type of regulation, the student teacher was classified in the pattern named after this type (retrospective active (1), prospective active (2), and passive regulation pattern (3)). In case the student teachers’ six learning experiences were equally spread over the three types of regulation, the pattern was named versatile (4). In case the learning experiences were spread over two types, the student teacher was categorised in a pattern that was named after this combination of types (combined prospective and retrospective active (5), combined prospective active and passive (6), and combined retrospective and passive regulation pattern (7)). With seven different regulation patterns, there were too many different change paths found within individual student teachers over time to categorise this change in a meaningful way as we did with their changes in learning orientations. Therefore, change in regulation patterns was only studied on the group level by calculating differences in spreading of patterns over time by means of Chi-square tests.

124

Chapter 5

5.4 Results 5.4.1 Student teachers’ change in regulation of learning and learning orientations For illustrative purposes and as a first indication of the effect of time on the different variables, the intraclass correlation, change in group mean and covariance stability will be explored. The intraclass correlation (ICC) as displayed in Table 5.1 shows how much of the total variance of each variable is variance between individuals. All these intraclass correlations were significant (p < .05) and varied from 15% (REG-RE) to 80% (LA-TP). This means that the other part is variance within individuals, or in other words variance over time. This shows that for all variables, it is worthwhile studying individual changes over time. Box-plot graphs were used to present group means at the three measurement occasions (see Figure 5.2). At first sight, comparison of these plots show linear as well as non-linear changes in group means over time. The covariance stability is described by exploring the correlations between the measurement occasions (see Figure 5.3). In general, the learning orientation scales score higher test-retest correlations than the regulation scales. The correlations of the learning orientation scores over time are rather high and the correlations of T1 with T2 in general do not differ a lot from T1 with T3, which resembles a compound symmetry model, indicating that not much variation in slopes can be expected. The correlations of the passiveness of regulation (REG-PS) show more evidence of increased variability over time, whereas for retrospectiveness of regulation (REG-RE) only a significant relation has been found between T2 and T3.

125

REG-PS

REG-RE

2

2

1

1

0

0

-1

-1

-2

-2 1

2

3

LC-PT 7 6 5 4 3 2 1

1

LC-SD

2

3

1

LC-CE

7

7

7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1 2

3

1

LC-IC

2

3

1

LC-TP 7

7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1 1

2

3

LA-PE

2

3

1

ER-AV 7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1

1 3

2

3

ER-PO

7

2

3

1 1

7

1

2

LA-VD

7

1

3

1

1 1

2

LA-PM

1 1

2

3

1

2

3

Figure 5.2. Box-plot graphs for all regulation and learning orientation variables over time (for abbreviations, see Table 5.1).

126

Chapter 5

REG-PS T1 T2 T3

T2 .55**

T3 .23 .21

T1 T2 T3

T2 .17

LC-SD T1 T2 T3

T2 .51**

T2 .65**

T3 .45** .53**

T1 T2 T3

T2 .60**

T2 .43**

T1 T2 T3

T3 .60** .81**

T1 T2 T3

T2 .82**

T3 .59** .74**

T1 T2 T3

T1 T2 T3

T2 .46**

T3 .61** .71**

T2 .49**

T3 .56** .78**

LA-VD T3 .76** .81**

T1 T2 T3

ER-AV T3 .36** .48**

T2 .67**

LA-PM

LA-TP

LA-PE T1 T2 T3

T3 -.02 .30*

LC-CE

LA-IC T1 T2 T3

LC-PT

REG-RE

T2 .67**

T3 .72** .74**

ER-PO T3 .45** .67**

T1 T2 T3

T2 .67**

T3 .55** .65**

* significant at p < .05, ** significant at p < .01

Figure 5.3. Overview of covariance stability for all regulation and learning orientation variables: Correlations between ratings of the same student teacher over time (for abbreviations, see Table 5.1). The relations between all variables have been explored by calculating correlations between all learning orientation scales and regulation dimensions across the three measurement occasions (see Table 5.2). Table 5.2 shows no significant relations (all r < .10) between the regulation dimensions and any of the learning orientation scales. For the correlations between the different learning orientations 26.7% was less than .10, 55.6% between .10 and .30 and only one correlation of > .50 (LA-IC with LA-TP). To conclude, the largest part of the variables show weak correlations (Cohen, 1992), which points out that the univariate multilevel analyses will be executed on set variables with relatively little overlap.

127

Table 5.2 Correlations between all variables across measurement occasions LC-SD LC-CE LA-PM LA-IC LA-TP LA-VD LA-PE ER-AV ER-PO LC-PT LC-SD LC-CE LA-PM

.29**

REG REG -PS -RE

-.08

.26**

-.04

-.11

.05

.22**

.00

.20**

.08

.08

-.36**

-.12

.09

-.02

.06

-.09

.02

-.10

.01

.08

.35**

.22**

.39**

.07

.27**

-.32**

.21**

.04

-.09

.21**

.15*

.35**

.20**

-.17*

.24**

-.02

.01

.67**

.41**

.28**

-.26**

.27**

.30**

-.30**

.02 .04

.01 .06

.07 .02

.12

-.10

-.15*

-.08

-.02

-.24**

.17*

.03

-.02

-.10

.08

-.06

-.11

.12

LA-IC LA-TP LA-VD LA-PE ER-AV ER-PO REG-PS

.08

* significant at p < .05, ** significant at p < .01

To answer the first research question regarding within-individual change in the two parts of self-regulated learning, first the results of the multilevel analyses will be described, followed by the results of change in orientations and patterns.

Individual change in scores on regulation of learning dimensions and learning orientation scales The results of the multilevel analyses are described in Table 5.3. In this table, only the accepted models are displayed, including the results of the deviance test for model testing, estimation of the intercept, growth estimations including p-levels. The predicted values for all individual student teachers, which are based on the various coefficients of the final models from Table 5.3, are also displayed in Figure 5.4.

128

(3)= 11.9** (1)= 5.0*

3 Quadratic growth model - fixed slope

1+No change over time model, with added predictor

1 No change over time model

3 Quadratic growth model - fixed slope

REG-RE

LC-PT

LC-SD

LC-CE

(1) 10.83***

(2)= 9.1* (1)=4.4*

3+ Quadratic growth model fixed slope with added predictor

1 No change over time model

4 linear growth model - random slopes

1+ No change over time model with added predictor

LA-VD

LA-PE

ER-AV

ER-PO

4.20 (.20)

1.81 (.06)

5.64 (.06)

4.57 (.17)

4.75 (.19)

3.33 (.13)

4.90 (.11)

5.50 (.08)

3.90 (.08)

4.80 (.09)

0.01 (.07)

-.10 (.06)

Intercept (SE)

0.08* (.04)

0.71*** (.18)

0.15* (.06)

-0.28* (.13)

0.46** (.16)

0.10* (.04)

Time (SE)

-0.26** (.09)

0.11† (.06)

-0.21** (.08)

Time2 (SE)

0.58* (.27)

-0.74*** (.22)

-0.49* (.25)

0.27* (.12)

PROG (SE)

1.10*** (.22)

0.14** (.05)

0.19*** (.05)

0.78*** (.15)

1.08*** (.19)

1.04* (.22)

0.74*** (.15)

0.35*** (.07)

0.33*** (.07)

0.22*** (.04)

0.05† (.03)

0.10*** (.03)

Intercept variance (SE)

0.04 (.02)

0.12* (.06)

Slope variance (SE)

Random effect coefficients

0.01 (.02)

-0.03 (.09)

Intercept x Slope (SE)

Interaction effect coeff

at p < .10, * significant at p < .05, ** significant at p < .01, ***significant at p < .001, The differences in deviances (-2 log likelihood) are distributed as a chi-square statistic and the difference in parameters with the previous accepted model defines the degrees of freedom (Peugh & Enders, 2005).

#

† significant

(1)= 156.9***

1+ No change over time model with added predictor

LA-TP

(1)= 34.2***

(2)= 6.7*

LA-IC

(1)= 75.1***

1 No change over time model

4 linear growth model, random intercepts, random slopes

LA-PM

(1)= 6.2*

(1)= 45.9***

(1)=6.6*

2 Linear growth model - fixed slope

Best fitting model

Model testing# χ 2(df)

REG-PS

Dependent variable

Fixed Effect Coefficients

Table 5.3 Overview of best fitting models for all variables, including results of model testing, and estimates for fixed, and random parts

Chapter 5

129

REG-RE

REG-PS 2

2

1

1

0

0

-1

-1

-2

-2 1

2

6 5 4 3 2 1 1

3

LC-PT 7

2

1

3

7

7

7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1

1

1 1

2

3

1

2

1

3

7

7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1

1 2

2

3

1

7

7

6

6

6

5

5

5

4

4

4

3

3

3

2

2

2

1 1

2

3

2

3

ER-PO

ER-AV

LA-PE 7

1

3

1 1

3

2

LA-VD

LA-TP

LA-IC 7

1

3

LA-PM

LC-CE

LC-SD

2

1 1

2

3

1

2

3

Figure 5.4. Predicted models for all regulation and learning orientation variables (for abbreviations, see Table 5.1).

130

Chapter 5

For six variables, the no-change-over-time model was the best fitting model, indicating that there was no individual growth on these variables. These stable variables all included ILTP-scales: the learning conceptions practising and testing (LC-PT) and strong self-determination in performance improvement (LC-SD) and the learning activity scales proactive, broad use of the mentor (LA-PM), actively relating theory and practice (LA-TP) and pupil-oriented evaluation criteria (LA-PE), as well as the preoccupation scale (ER-PO). This means that although student teachers had different values on these variables on T1, these values did not significantly change over time. The predicted values of these variables show flat lines in Figure 5.4. For three variables a linear growth model fitted the data best. The passiveness of regulation (REG-PS) increased for an average student teacher from -0.10 to 0.10 from the first measurement occasion (T1) to the last (six months later). With the standard deviation of the intercept scores as a reference, this means an increase of 0.63 SD. For the variables independent search for conceptual information (LA-IC) and the avoidance scale (ER-AV) a linear growth was found, but also individual differences in the growth rate were found to be significant. This indicates that the general trend is an increase in values, but some student teachers showed more increase than others (who might even decrease in scores over time). This can also be recognised in Figure 5.4 by the non-parallel lines of the predicted values for these variables. The independent search for conceptual information (LA-IC) increased for an average student teacher with an average slope from 3.33 to 3.63 (an increase of 0.29 SD), and the avoidance scale (ER-AV) increased from 1.81 to 1.97 for an average student teacher with an average slope (an increase of 0.38 SD). The retrospectiveness of regulation (REG-RE) and developing views/ideas through discussion (LA-VD) both showed a quadratic growth by first increasing and then decreasing. The retrospectiveness of regulation increased from 0.01 at T1 to 0.26 at T2, which is an increase of 1.12 standard deviation. This points out that an average student teacher at T2 scores as high as a high scoring student teacher at T1. On T3, the score of an average student teacher on retrospectiveness of regulation (REG-RE) is decreased to 0.09, which is still an increase of 0.25 standard deviation compared with T1. The intercept score of developing views/ideas through discussion (LA-VD) increased for the job-track student teachers (PROG = 0) first from 4.57 to 5.02 (0.47 SD) and then decreased to 4.95, which is nevertheless an increase of 0.40 standard deviation compared with the score at T1. Raising consciousness under external control (LC-CE) also showed a 131

quadratic growth pattern, but then by first decreasing from 5.50 to 5.33 (0.29 SD) and then increasing to 5.38, which is still 0.20 SD lower than the value at T1.

Individual change in learning orientations and regulation patterns To answer the first research question about individual growth in terms of change in patterns, first differences in distribution of the different learning orientation and regulation patterns over time were explored. As can be seen in Table 5.4, the frequencies of the seven regulation patterns were different on the three measurement occasions (χ2(12) = 23.53, p = .02). Especially noticeable is the high number of student teachers with a passive regulation pattern on T3, the low number of student teachers showing versatile regulation on T2, the general decrease in student teachers with a combined prospective and retrospective active regulation pattern, and the increase from measurement occasion 1 to measurement occasion 2 in the number of student teachers with a combined active retrospective with passive regulation pattern. Table 5.4 Distribution of regulation patterns over time (frequencies) Regulation pattern

T1

T2

T3

Total

1 Retrospective active

3 (4.0%)

6 (8.5%)

2 (2.9%)

11 (5.1%)

2 Prospective active

5 (6.7%)

8 (11.3%)

5 (7.2%)

18 (8.4%)

3 Passive

5 (6.7%)

5 (7.0%)

11 (15.9%)

21 (9.8%)

4 Versatile

10 (13.3%)

2 (2.8%)

7 (10.1%)

19 (8.8%)

5 Combined prospective & retrospective active

17 (22.7%)

13 (18.3%)

8 (11.6%)

38 (17.7%)

6 Combined prospective active & passive

9 (12.0%)

22 (31.0%)

14 (20.3%)

45 (20.9%)

7 Combined retrospective active & passive

26 (34.7%)

15 (21.1%)

22 (31.9%)

63 (29.3%)

Total

75 (100%)

71 (100%)

69 (100%)

215 (100%)

The distribution of learning orientation over time, was not significantly different (χ2(6) = 4.38, p = .63), see Table 5.5. However, a closer look at the individual changes in learning orientation (see last column of Table 5.6) show that in total 37.0 % of the student teachers maintained the same learning orientations over time, where 34.2 % changed positively towards more

132

Chapter 5

independent meaning-oriented, and 19.2% developed towards more survivaloriented learning. Table 5.5 Distribution of learning orientations over time for student teachers from the interntrack and job-track (frequencies) T1 Learning orientation Independent meaningoriented Dependent meaningoriented Reproductionoriented

Survivaloriented

Intern N=45

T2 Job N=33

17 (21.8%) 10 (22.2%)

7 (21.2%)

23 (29.5%) 9 (20.0%)

14 (42.4%)

16 (20.5%) 12 (26.7%)

4 (12.1%)

22 (28.2%) 14 (31.1%)

8 (24.2%)

Intern N=44

T3 Job N=30

18 (24.3%) 5 (11.4%)

13 (43.3%)

19 (25.7%) 12 (27.3%

7 (23.3%)

13 (17.6%) 11 (25.0%)

2 (6.7%)

24 (32.4%) 16 (36.4%)

8 (26.7%)

Total

Intern N=41

Job N=28

20 (29.0%) 10 (24.4%)

10 (35.7%)

17 (24.6%) 7 (17.1%)

10 (35.7%)

18 (26.1%) 17 (41.5%)

1 (17.6%)

14 (20.3%) 7 (17.1%)

7 (25.0%)

Intern N=130

Job N=91

55 (24.9%) 25 (19.2%)

30 (33.0%)

59 (26.7%) 28 (21.5%)

31 (34.1%)

47 (21.3%) 40 (30.8%)

7 (7.7%)

60 (27.1%) 37 (28.5%)

23 (25.3%)

Table 5.6 Changes in learning orientation for student teachers from the intern-track and jobtrack (frequencies) Type of change in learning orientation

Internship

Job

Total

No change

12 (27.9 %)

15 (50.0%)

27 (37.0 %)

Survival

2 (4.7 %)

5 (16.7 %)

7 (9.6 %)

Reproduction

4 (9.3 %)

1 (3.3 %)

5 (6.8 %)

Dependent meaning

3 (7.0 %)

6 (20.0 %)

9 (12.3 %)

Independent meaning

3 (7.0 %)

3 (10.0 %)

6 (8.2 %)

Positive change

17 (39.5 %)

8 (26.7%)

25 (34.2 %)

Negative change

11 (25.6 %)

3 (10.0 %)

14 (19.2 %)

Up and down

0 (0.0%)

3 (10.0 %)

3 (4.1 %)

Down and up

3 (7.0 %)

1 (3.3 %)

4 (5.5 %)

43 (100 %)

30 (100 %)

73 (100%)

Total

133

To summarise, the results regarding within-individual change showed that contrary to our expectations, student teachers became more passive in their regulation throughout the programme. Furthermore, the amount of retrospectiveness of regulation (REG-RE) showed a peak on T2. The majority of the scores on the learning orientation scales remained stable over time. Two scales (LA-IC and LA-VD) which are positively related to meaning-oriented learning increased over time, however some other indicators for independent meaning-oriented learning (LC-SD and LA-TP) did not. Furthermore the decrease on LC-CE showed that at the end of the programme student teachers agree less on the view that learning to teach means raising consciousness under external control. No differences were found in the distribution of learning orientations over time on the group level. On the individual level, the number of student teachers changing towards independent meaning-oriented learning was larger than the group changing away from it.

5.4.2 Differences between programme tracks The results with respect to the between-individual differences (research question 2), will be presented by first showing the outcomes of the addition of the predictor variable PROG to all accepted multilevel models. In addition, differences between the tracks in terms of distribution of regulation patterns and (change in) learning orientations will be shown.

Differences in scores on regulation of learning dimensions and learning orientation scales For two variables a random slope model was accepted (LA-IC and ER-AV), which means that only for these two variables the effect of the programme track on the growth rate could be tested. It turned out that for these variables the predictor PROG did not explain variance in intercept or slope. For the other variables, we could only investigate whether the values at T1 (intercepts) differed significantly between the groups. For four variables, the addition of the programme as a predictor showed a better fit by explaining differences in intercepts (marked with a “+”sign in Table 5.3). The coefficient of the predictor PROG shows the effect of being an intern on the predicted values. The average value on T1 of an intern on the learning conception practising and testing (LCPT) was 5.07, compared to 4.80 of a student teachers with a paid job (t(80.16) = 2.28, p = .03). The track of the programme explained 6.33% of the variance on the intercept score. On average, interns also scored higher on the 134

Chapter 5

scale preoccupation (ER-PO) (M = 4.78), compared to the student teachers with a job (M = 4.20, t(81.59) = 2.14, p = .04, R2 = 5.59%). On the other hand, job-track student teachers scored higher on actively relating theory and practice (LA-TP) (Mjob = 4.75, Mintern = 4.26, t(79.77) = -1.99, p = .05, R2 = 4.54%). Also on the variable developing views/ideas through discussion (LA-VD) the job-track student teachers scored on average higher (M = 4.57) than the interns (M = 4.01, t(78.73) = -3.39, p = .001, R2 = 12.48%).

Differences in regulation patterns and learning orientations The student teachers from the two programme tracks did not differ in their distribution over the regulation patterns (χ2(6) = 5.42, p = .49). In Table 5.5, the distribution over the learning orientations for the two programmes is displayed. The distribution over the learning orientations were on T2 and T3 significantly different for the student teachers from the two different programmes (T1: χ2(3) = 5.54, p = .14; T2: χ2(3) = 11.53, p = .009; T3: χ2(3) = 12.76, p = .005). This shows that the differences between the groups increased over time: the jobtrack showed more meaning-oriented learning at the end of the programme, and the intern-track more reproduction-oriented learning. Also for the change in learning orientation differences between the job-track and intern-track in development are significant. In Table 5.6 it can be seen that job-track student teachers more often maintained the same learning orientation, where interns changed more in both directions (χ2(4) = 10.15, p = .04). In summary, the programme track had no effect on the scores on the regulation dimensions and regulation patterns. However, the job-track student teachers have higher scores on two variables related to independent meaningoriented learning (LA-TP and LA-VD), whereas interns scored higher on variables less related to this type of learning. Furthermore, the job-track showed more meaning-oriented learning orientations at T2 and T3. These results are in contrast with our expectations regarding the second research question that interns would in general develop a more preferable learning orientation than students with a paid job.

5.5 Conclusion and discussion Two research questions were posed related to student teachers’ development in self-regulated learning in two tracks of a dual learning programme. Below, the results related to these questions will be discussed.

135

The first research question focused on student teachers’ change in regulation of learning and learning orientations throughout a one-year dual teacher education programme. The results of the longitudinal multilevel analyses showed that both dynamic aspects of regulation of learning changed over time, showing an increase in passiveness and retrospectiveness of regulation. The changes in regulation patterns were significant over time as well, reflecting an increase in the frequency of student teachers with a passive regulation pattern, and a decrease in the combined prospective and retrospective active regulation pattern. These results are not in line with our expectations that student teachers would develop towards self-regulated learners to be prepared for lifelong learning (Kremer-Hayon & Tillema, 1999). However, the increase of passive regulation does not imply that student teachers lost their ability to regulate their learning in an active way. The puzzling question is why student teachers became less triggered to use their active regulation skills over time. One explanation for this could be that student teachers have a smaller need to direct their learning actively at the end of the programme, since they become more competent as a teacher. It is possible that they become more satisfied about their teaching skills, might have fewer bad lessons and thus develop more in terms of getting routines instead of continuously developing new teaching skills. This consolidation in practice has also been described as one in the teachers’ professional life phases by Huberman (1989). After the career entry that goes along with feelings of survival and practice shock, but also a sharp learning curve stimulated by the challenge of discovery, the second phase is characterised as a stabilisation period. Teachers in this stage get a growing sense of mastery and comfort. However, Huberman indicates that this phase mostly occurs three to five years after entering the profession. Our results give rise to the idea that such a phase might also occur earlier in the teaching career, reflecting a desire of teachers who are starting out to consolidate what they have learned in the first hectic period of starting to teach. This explanation is also in line with the results related to the change in retrospectiveness of regulation. These results showed that the student teachers were most proactive in their regulation, for example by planning their learning at T1. The fact that the explicit planning of learning is more often seen at the beginning of the programme might reflect the increased attention and high desire to learn and to reach specific learning goals. From studies into experienced teachers’ regulation of learning (Van Eekelen, Boshuizen, & 136

Chapter 5

Vermunt, 2005) and workplace learning in general (Eraut, 2004), we know that unplanned and reactive learning is a major part of learning in informal learning settings. The expanding amount of lessons and responsibilities at school throughout the year might therefore be one of the factors that caused the increase of retrospective regulation characterised by more reactive and less planned learning at T2. At the end of the programme, student teachers’ regulation might become again more focused on prospectively directing their learning to reach the end terms of the teacher education programme. It is important to realise that the existence of the second dimension discerning prospective from retrospective regulation, showed that for student teachers, unplanned learning is not the same as lack of regulation. When showing retrospective active regulation, student teachers might not have planned their learning beforehand, but still were able to actively regulate their learning by monitoring, evaluating and reflecting on it in a more advanced way. The results of increased retrospectiveness of regulation thus also show greater attention to these phases of regulation throughout the programme. Multilevel analyses of the learning orientation scales showed for six scales no change over time. Only the increase of independent search for conceptual information and developing views/ideas through discussion supports our expectation that student teachers developed towards more independent meaning-oriented learning. On the other hand, the scales strong self-determination in performance improvement and actively relating theory and practice, which are also related to independent meaning-oriented learning, did not change over time. From the analysis of individual changes in learning orientations we can conclude that some of the student teachers changed towards a more meaning-oriented way of learning, but that the majority of student teachers did not show the expected development in learning orientation that is assumed to be necessary in becoming an expert teacher (Hagger et al., 2008; Oosterheert, 2001). This is not in line with the increase of meaning-oriented learning for student teachers’ academic learning orientation as was found by Donche and Van Petegem (2009) in their two-year study. This might indicate that it is easier for student teachers to develop towards meaning-oriented academic learning than to develop meaning-oriented learning in a dual learning programme in which academic learning and learning in practice are supposed to be intertwined, or that more time is needed for such a change.

137

An explanation for the lack of change towards independent meaningoriented learning during learning to teach might be that such a change does not occur by merely integrating more theory into practice and by trying to understand these theories. It also involves becoming more uncertain of one’s own perception and thinking and being open to other perspectives (Oosterheert, 2001). According to Oosterheert, this might be hard to reach for student teachers since they have to temporarily let go their control of their learning while they intuitively strive towards more control of their teaching. Illeris (2007) has stated that this barrier to learning might by caused by a type of psychological defence “to protect the individual against learning which, for one reason or the other, can be threatening, limiting or in some way places a strain on maintaining mental balance” (Illeris, 2007, p.160). Such defence mechanisms typically take place when situations are practically and psychologically overwhelming, such as teaching situations might be for inexperienced teachers. These defence mechanisms might become stronger when practice becomes more demanding, as it is for interns after changing schools. This could be an explanation for the change away from independent meaning-oriented learning, as demonstrated by some student teachers. The second research question was focused on the between-individual differences in regulation of learning and learning orientations of student teachers from the different types of programme they followed (job-track and intern-track). The expected negative effect on the development of selfregulated learning of the time-intensive job-programme could not be confirmed. There were no differences in regulation of learning between the two tracks. When differences in learning orientations were found, the student teachers with a job showed a more preferable way of learning. They scored higher on actively relating theory and practice and developing views and ideas through discussion, and lower on the learning conception that practising and testing is sufficient for learning to teach and they were less preoccupied with bad teaching experiences. Furthermore, student teachers with a job showed more often a meaning-oriented way of learning orientation than interns who were especially more often reproduction-oriented in their learning. Explanations for these results are only speculative. Oosterheert et al. (2002) already found that learning orientation is not related to teaching experience , so it is not likely to expect that student teachers with a job are already a step ahead in their development towards an independent meaningoriented way of learning, especially as the differences turned out to increase 138

Chapter 5

throughout the programme. Furthermore, many researchers have shown that practice alone is not enough to develop necessary learning skills and to become an expert teacher (Darling-Hammond, 1999; Hagger et al., 2008). Explanations for the differences found might point to the richness of the experience of student teachers with a job and the responsibilities they have and take. In an internship, student teachers might especially in the first semester not feel like a regular teacher, but more like an observer while someone else is responsible for the pupils’ learning. Therefore, they might engage less in discussions with their “not-real” colleagues and not feel an urgent need to actively search for theoretical background to their teaching by themselves. Another interesting finding is that although in general job-track student teachers have a more preferable way of learning, their learning orientations are more stable than those of the interns. Interns change more from learning orientation, however, both in a negative and a positive way. This could be related to the change of school and thus also of supervisor after the first semester which occurs in the intern-track, and not in the job-track. It would be interesting to study to what extent student teachers’ change towards a more or less preferable way of learning is related to the mainstream learning conceptions of their supervisor at school and the student teachers’ perception of their school as a supporting learning environment. From other studies we know that supervising teachers differ in their roles and their learning conceptions (Zanting, Verloop, & Vermunt, 2001), which might influence student teachers’ learning and regulation activities. The reason to study both parts of SRL, was that learning orientations are expected to influence how students actually learn (Brownlee, Purdie, & Boulton-Lewis, 2003) and the other way around, students’ situation-specific learning experiences are regarded as accumulating towards their general learning orientations (Makinen & Olkinuora, 2004). Nevertheless, these aspects have been mostly studied separately. When learning orientations measured by general questionnaire and regulation of specific learning experiences were included in the same study, often low correlations have been found (Veenman, 2005). Although meaning-oriented learning is often associated with preferences for self-regulation learning (Vermunt, 1998; Vermunt & Vermetten, 2004), also in our study no significant correlations were found between the regulation of learning dimensions and learning orientation scales. More research is needed to understand the exact relation between these two components of SRL. 139

In conclusion, although the literature has described the importance of SRL as a means and an end in teacher education, this study found little evidence that student teachers become more self-regulating throughout the teacher education programme. The results of this study confirmed the expectation of Oosterheert (2001) that student teachers’ ways of learning will not spontaneously change towards active, independent and meaning-oriented learning. This shows that increasing opportunities for directing one’s own learning (Kremer-Hayon & Tillema, 1999) might be a necessary, but not a sufficient condition to increase student teachers’ conceptions and skills to become a self-regulated lifelong learner. Since this study took place at only one teacher education institute, further studies in other institutes are necessary to confirm these results. If this is the case, the next intriguing question to answer would be why so many student teachers who are supposed to become experts in learning, did not develop towards an actively regulated and meaningoriented way of learning, which ironically is also the type of learning they are nowadays expected to promote in their own pupils (Kremer-Hayon & Tillema, 1999). Some possible explanations have been given, though more research is needed to discover the reasons why student teachers prefer a certain learning orientation and certain types of regulation. This information could be used to develop together with teacher educators interventions for the heterogeneous group of student teachers to prepare them for active lifelong learning.

140

existing

patterns categorised

ideas

selection

tools

lack

programme

combination

presented

below

scores identified

concept

report

individual

types

learning report

students

behaviour

defined

learner

srl direction variance

important reference

relevance

data reactive

collected reliable

research information

seems

professional

competence become

meaning-oriented

dissertation

tracks

nevertheless

niemivirta

structured distinction multiple

described

dynamic

explain

limited

prospective

experiences

small-scale

steer

oosterheert

alternative

results

interpretation

retrospective engaging

aim

instruments

improve

acknowledge

concrete

variables

insights

focused

expected

dimensions

considering

job-track

sample

active

primarily

external ability

already

stable

specific

assessment

student teachers'

increased

implications

education

choosing

independent

growth

university

chapter

integrated

working

reflection quality

iltp

following

suggestions

validity

personal framework assumed scales

factors

theories large-scale analysis

tested

development

academic

opportunity

caused

project

empirical

netherlands

occasion

dual

aspects

portfolio

conclusion

passive

reproduction-oriented

influence

different

detail

teacher

experienced

means

action

phase

literature

relation

skills

larger

contribute

completely

understanding

deliberate

variety

higher

description

sometimes

discriminating

frame

learning

focus

further

variation

measurement

assumption

asked

environment

correspondence

previous

approach

vermunt

goals

preferences

indicates

responsibility

model

orientations

extent

categories

conditions

participation

strategies

context

stimulate

regulation

outcomes

coding

versus

conceptual

included

role evaluated

besides

regarding workplace

answers

self-regulated

6 General conclusions and discussion In the General Introduction, we gave several arguments for self-regulated learning (SRL) as a means and end in pre-service teacher education programmes. As Fallon (2008, p. 837) has concluded, “the field of teacher education and teacher learning is deep and rich in normative and logical reasoning, but shallow in empirical knowledge”. Therefore, the aim of this study was to contribute, with empirical data, to the understanding of the nature and development of student teachers’ SRL in a dual learning programme. To get a good overview of both the stable and dynamic part of SRL, we have studied both student teachers’ general learning orientations and the more context-specific regulation activities of student teachers simultaneously. In this dissertation we viewed and assessed learning orientations as the combination of student teachers’ learning conception and preferences for certain learning and regulation activities. Student teachers’ regulation of learning was defined as the combination of regulation activities as performed in concrete learning experiences. Our assumption was that the improved understanding of the nature and development of SRL is crucial for teacher educators to design their instruction to promote the development of student teachers as active learners. In this final chapter we will not only give answers to the questions that were posed, but also discuss the main results in terms of the existing literature and describe the implications and suggestions for practice.

6.1 Questions with answers The main research question of this dissertation, as posed in the General Introduction, was: What is the nature and development of student teachers’ selfregulated learning in a dual learning programme? To answer this research question, two empirical studies have been carried out with the following more specific research questions:

143

1. What is the nature of student teachers’ regulation of learning across different learning experiences in the two contexts of a dual learning programme? 2. What are differences in student teachers’ regulation patterns and learning conceptions in a dual learning programme and how are these related to each other? 3. To what extent is a structured multiple event question log a valid and reliable instrument for measuring and discriminating different qualities of student teachers’ regulation of learning in a dual learning programme? 4. How do student teachers’ regulation of learning and learning orientations change throughout a one-year dual learning programme and what are differences between the two programme tracks? Below, we will integrate the results of the two empirical studies to answer these specific questions.

6.1.1 Nature of student teachers’ regulation of learning The first question posed concerned the nature of student teachers’ regulation of learning in a dual learning programme. Content analysis of 133 reported learning experiences in open question logs on eight aspects of regulation of learning, resulted in a rich description of more than 50 categories describing the large variety in student teachers’ regulation of learning, as described in Chapter 2. The results of a multiple correspondence analysis on these categorised learning experiences revealed the underlying structure describing the nature of student teachers’ regulation of learning in a dual learning programme. Regulation of student teachers’ learning experiences appeared to differ on two dimensions: passive versus active regulation and prospective versus retrospective regulation. This two-dimensional structure was validated in the second study, described in Chapter 4, in which 81 student teachers were included from whom 1292 learning experiences on three measurement occasions throughout the teacher education programme were collected. The only difference between the spreading of the data from both studies on the two dimensions was that in the second study there was less variation in scores on the second dimension on the passive side of the first dimension. Therefore, in the larger scale study only three types of regulation were discerned, instead of the four types of regulation based on all four quadrants as described in Chapter 2. In these three types of regulation (see Table 6.1), no distinction has 144

Chapter 6

been made between prospective and retrospective passive regulation. Taken together, with a distinction in three types of regulation based on the twodimensional structure as described above, the nature of regulation of student teachers’ learning experiences can be conceptualised. Table 6.1 Overview of three types of student teachers’ regulation of learning Type of regulation

Description

Passive regulation

Lack of regulation or reliance on external regulation in all phases of the learning process.

Active prospective regulation

Proactive way of regulation. The focus of the regulation activities is on the forethought phase of the learning process. Learning experiences are often planned, well-thought goals are set, and arguments for the choice of a certain learning strategy are given. The regulation of monitoring, reflection, and evaluation is more superficial.

Active retrospective regulation

Reactive, more reflective way of regulation. Learning experiences are often unplanned, so no goal-setting or deliberate thinking about learning strategy and self-efficacy takes place. Active retrospective regulation is characterised by deliberate, deep ways of monitoring, evaluation and reflection, not only focused on what is learned, but also on how it is learned.

The interpretation of the second dimension implied that learning experiences starting as reactive and non-deliberate and lacking a pre-set learning goal, can still involve deliberate, active regulation activities, though in a more retrospective way. These results showed that goal-setting and planning is not a necessary condition to show active regulation of learning in a dual learning programme. The small-scale study also focused on the difference in nature of regulation between the two contexts of a dual learning programme, namely the practice school and the university. The results showed that learning experiences that took place while learning in practice were significantly more actively regulated than learning experiences that took place while learning at the university. These results suggest that although learning at the workplace is often regarded as being less deliberate and more reactive in nature (Tynjälä, 2008), the context of practice triggered more active regulation than the university context.

6.1.2 Individual differences in self-regulated learning The insights into the nature of SRL on the level of the learning experience as described above were used to describe the variety in SRL on the level of the 145

individual student teacher. To describe this variety, both parts of SRL were taken into account: student teachers’ regulation of learning described in patterns and student teachers’ learning conceptions (Chapter 3) and learning orientations (Chapter 4 and 5), including the relation between them. First the outcomes of the variety in regulation of learning will be described, then the variety in learning conceptions and orientations, and finally the relation between them. Analysis of the data of the small-scale study as described in Chapter 3, showed that student teachers employed multiple types of regulation in learning to teach. The six learning experiences of every student teacher were pictured in one of the quadrants of the two-dimensional structure to describe their individual regulation pattern. Five student teachers were categorised as having a versatile regulation pattern and one of the student teachers had a clearly passive regulation pattern. The rest of the student teachers could be categorised as being active, prospective or retrospective regulators, meaning that at least a substantive part of their learning experiences was actively regulated. In the large-scale study, the three types of regulation (see Table 6.1) were used to categorise the data, instead of the four quadrants of the two dimensions. In this data set, two additional regulation patterns were found compared to the small-scale study. Some of the regulation patterns classified as one pattern in the first study were distinguished as different patterns in the larger sample. From the resulting seven patterns, four patterns were combination patterns, meaning that multiple types of regulation were included. More than three-quarters of the student teachers were classified as showing such a combination pattern. In conclusion, most student teachers showed multiple types of regulation while learning in a dual learning programme, and almost all student teachers showed at least once within their six learning experiences an active way of regulating their learning. In the small-scale study as reported in Chapter 3, interviews were used to study learning conceptions. For this study we used the concept of learning conceptions instead of learning orientation, since we only asked student teachers on a very global level for their preferences for learning and regulation activities. To classify student teachers’ conceptions, we used the main distinction between meaning-oriented and reproduction-oriented learning as described by Oosterheert and Vermunt (2001). In our sample, only a small number of student teachers (eight, i.e. 29%) showed a meaning-oriented learning conception. In the larger-scale study, the Inventory Learning to Teach 146

Chapter 6

Process (ILTP) was used to measure student teachers’ learning orientations. The student teachers were equally distributed over the four learning orientations as described by Oosterheert, Vermunt, and Veenstra (2002). Cross-tabs were made to calculate in the small-scale study how student teachers’ learning conceptions were related to their regulation patterns. The concepts of consonance and dissonance were applied to interpret these relations; however, the results were inconclusive. In this study only three (14%) student teachers showed a theoretically consonant pattern of meaningoriented learning conception and active regulation or reproduction-oriented learning conception with passive regulation (Beishuizen, Stoutjesdijk, & Van Putten, 1994; Vermunt, 1998). On the other hand, the rest of the combinations could not be directly classified as theoretically dissonant. This was partly caused by the fact that current regulation theories do not include the newlyfound dimension prospective versus retrospective regulation. Although it has often been stated that for a meaning-oriented learning approach, active regulation is needed (e.g., Biggs, 1988), our data showed that the opposite does not have to be the case. In particular, active prospective regulation could very well go together with a reproduction-oriented learning conception in which student teachers subsequently try to improve their learning step-by-step in a specific direction without questioning their frame of reference. In the larger data set as described in Chapter 4, also no significant relation was found between student teachers’ learning orientations and their aggregated scores on the two regulation dimensions. These results show that in a dual teacher education programme, multiple combinations of both parts of SRL may exist.

6.1.3 Measuring self-regulated learning The large variety in regulation of learning within student teachers, as described in Chapter 3, showed that exploring one learning experience is not enough to give a valid estimation of student teachers’ regulation of learning. To be able to collect longitudinal data in a larger sample, a less labourintensive way than the open question log was needed for the student teachers as well as the researchers to collect data of multiple events. Therefore, a structured version of the Learning Report was developed based on the descriptions from the answers on the open questions and evaluated with the following research question: To what extent is a structured multiple event question log a valid and reliable instrument for measuring and discriminating different qualities of student teachers’ regulation of learning in a dual learning programme? The results 147

were described in Chapter 4 and showed that although in the Structured Learning Report the rich descriptions of the open question version were absent, the reliability and construct validity did not decrease. The Structured Learning Report was suitable for measuring different types of learning experiences from different contexts. Furthermore, combining multiple learning experiences into a regulation pattern could discriminate different qualities of student teachers’ regulation of learning at a certain moment. Also data was collected to study the relation of the outcomes of the Structured Learning Report with the ILTP and with the student teachers’ final assessment score. No significant relations were found between these instruments. From previous studies we know that often no relation is found between aptitude instruments such as the ILTP and event measurement, such as the Structured Learning Report (Veenman, 2005). Furthermore, the only quantitative measurement available to measure learning outcomes of the student teachers was a rubrics scoring system, used by teacher educators to give a general description of student teachers’ competence level on a four-point scale. The low discriminative power and the fact that the instrument was still in its pilot phase might have influenced the lack of relations that were found. Future studies will be necessary to assess the predictive and concurrent validity of this instrument.

6.1.4 Development of self-regulated learning Our small-scale study with a cross-sectional design showed that student teachers at the end of the programme less frequently showed a meaningoriented learning conception than student teachers in the first part of the programme. The sample of this study was too small to see any trends in changes in regulation patterns. To measure change within individuals, a longitudinal study was set up including three measurement occasions. This study focused on how regulation of learning and learning orientations of student teachers from two tracks (i.e., job-track and intern-track) changed throughout a dual teacher education programme. The results of this study were described in Chapter 5. About 90% of the student teachers (N = 81) of one institute in The Netherlands participated in this study. Their learning orientations were measured by means of the ILTP and their regulation of learning of six learning experiences per occasion by means of the Structured Learning Report.

148

Chapter 6

Multilevel analysis and analysis of the distribution of regulation patterns showed that student teachers changed in their regulation of learning over time. However, the direction of change was not in line with our hypothesis that student teachers would become more active regulators during the programme. Throughout the programme, the passiveness of regulation increased. Regarding retrospectiveness of regulation of learning, the results showed a quadratic growth model with the lowest value at the first measurement occasion (three months after the start of the programme), and the highest value at the second measurement occasion (three months later), fitted these data best. At the end of the programme, the retrospectiveness decreased again, but still ended a bit higher than at the first measurement occasion. This shows that in the first part of the programme student teachers were most prospective in their learning, for example by setting learning goals and planning their learning. At the second measurement occasion student teachers regulated most retrospectively by focusing their regulation on monitoring, evaluating and reflecting on their learning. Multilevel analyses of the learning orientation scales showed for six of the ten scales no change over time. Two scales related to an independent meaning-oriented way of learning increased over time: independent search for conceptual information and the scale developing views/ideas through discussion. However, two other scales related to this type of learning did not change over time (strong self-determination in performance improvement and actively integrating theory and practice). Analysis showed that there were no significant differences in the distribution of learning orientations over time. The portion of survival and reproduction-oriented student teachers was quite consistent over time, around 50%. Individual lines of development, showed that more than one-third of the student teachers did not change from learning orientation, about one-third of the student teachers changed towards an independent meaning-oriented way of learning, and a smaller part developed away from the independent meaning-oriented learning orientation. In both the multilevel analysis and analysis of patterns, the effect of the type of programme (job-track and intern-track) on the development was also studied. The hypothesised negative effect on the development of self-regulated and meaning-oriented learning of the time-intensive job-track could not be confirmed. On the contrary, comparisons between the two groups showed that when differences were found, in general the student teachers with a job showed a more preferable way of learning. They scored higher on the learning 149

orientations scales “actively relating theory and practice” and “developing views and ideas through discussion”, and lower on the learning conception that practicing and testing is sufficient for learning to teach and they were less preoccupied with bad teaching experiences. In conclusion, the results of this study showed that in this dual learning programme, most of the student teachers did not develop towards active and independent meaning-oriented learners, that is assumed to be needed to continue learning after graduation (Hagger, Burn, Mutton, & Brindley, 2008; Oosterheert, 2001).

6.2 Questions without answers Some questions did not result in definite answers. Sometimes the answers were incomplete, inconclusive and sometimes they were very clear, but hard to interpret. Below, we will elaborate on this by giving our ideas and formulating hypotheses to direct our thinking about these questions.

6.2.1 What is the relation between the two parts of self-regulated learning? One of the main questions that could not be answered empirically was how student teachers’ learning orientations and their regulation of learning in concrete learning experiences were related to each other. These two aspects have for decades been regarded as being part of the same concept, although they are mostly studied separately (Dinsmore, Alexander, & Loughlin, 2008). In our studies, the stable and dynamic parts of SRL were studied simultaneously, though from different perspectives with different types of measurements. Previous studies have shown that often low correlations have been found between general aptitude measurements and event measurement (Veenman, 2005). These results have often led to questions concerning the validity of both types of instruments, instead of more fundamental questions about how these two parts are related to each other. Considering that every learning experience happens in a certain context that influences the type of regulation of the learning processes, we expected that by including multiple events we could rule out this influence to some extent. In this way we hoped to be able to find clearer relationships between the two parts and to bring both research traditions closer to each other. The small-scale study, as described in Chapter 3, indeed showed that almost all student teachers employed multiple types of regulation in their learning 150

Chapter 6

experiences. Therefore, regulation of some learning experiences was in line with their learning conception and others were not. Although student teachers will probably never regulate all their learning completely coherently with their learning orientation, we expected to find differences between student teachers with dissimilar learning orientations in their average way of regulation across learning experiences. However, in the large-scale study we also did not find a relation between student teachers’ general learning orientations and their regulation of learning. Another suggested explanation for lack of relations between these two parts is that a new learning environment may cause dissonant patterns, since the new environment might demand different learning and regulation activities than fit in the student’s “old” learning orientation (Lindblom-Ylänne & Lonka, 2000). However, these frictions are mostly only temporal and also in our study at the end of the programme no relation was found. One of the problems for interpreting the different combinations between learning orientations and types of regulation on a theoretical basis is that we have used a larger variety of combinations between the two parts than in previous studies. In the literature, only the combination of self-regulation and a meaning-directed learning approach, or the combination of external regulation and a reproduction-oriented learning approach, have been described (Beishuizen et al., 1994; Vermunt, 1998). By using the conceptual framework of Oosterheert and Vermunt (2001), we also made the distinction between dependent and independent meaning-oriented learning and included also survival-oriented learning. Furthermore, our own study revealed the dimension prospective versus retrospective regulation, which was also not identified in previous studies. In Chapter 3, we showed how active prospective regulation could fit as well in a meaning-oriented way of learning as in a reproductive learning orientation. This suggests that the relations between learning orientation and regulation of learning might be less straightforward than has been assumed before. Previous studies have also suggested that for adopting a new learning approach first external regulation is needed before students are able to fully adopt it (Vermunt, 1996). This indicates that the application of certain regulation types might also be related to a phase in the change process from one learning orientation to another, instead of to a specific learning orientation by itself. In conclusion, the lack of relations found in several studies suggests that both parts of SRL might be more complementary than overlapping parts of the same concept. The fact that they 151

seem to measure different aspects of the concept of SRL, shows the value of studying both parts simultaneously to get an integrated picture of the quality of student teachers’ SRL. More research is needed to find out why student teachers show specific combinations of learning orientations and types of regulation and under what conditions. Hopefully this will contribute to a more integrative framework of SRL as Brown (1987) hoped for more than two decades ago.

6.2.2 Does self-regulated learning make a difference? The main assumption for starting this research project was that SRL is an important factor in learning to teach. The importance of SRL for academic learning, as well as for expertise development while learning at the workplace has often been mentioned in the literature (Beijaard, Korthagen, & Verloop, 2007; Bolhuis, 2003; Clarke & Hollingsworth, 2002; Hammerness et al., 2005). Therefore, proving this assumption was not a specific research aim of this project. Nevertheless, the most asked question by far about this research project by practitioners, student teachers and also researchers is: Will these self-regulating student teachers also become better teachers? Our study could not find a relation between student teachers’ regulation of learning and their final assessment score measured by a four-point scale rubrics scoring system. This means that active or retrospective regulators did not leave the programme with a higher or lower final assessment score describing their teaching competences than passive or prospective regulators. The fact that this question could not be answered in the relatively small sample and the type of measurement used makes the question still intriguing and relevant. In particular, when teacher educators need to be convinced to change their teaching practice to improve more SRL in student teachers, we need an answer as to what extent and for what SRL makes a difference. In the literature, as described in Chapter 1, the following assumptions regarding the relevance of SRL were described: student teachers and experienced teachers engaging in SRL are expected to learn successfully in, and from a combination of, theory and practice (Mutton, Burn, & Hagger, 2010), to be able to continue lifelong learning after graduation (Hagger et al., 2008); to handle the challenges of the complex and dynamic learning environment of everyday practice (Hammerness et al., 2005), and to have more agency in external change processes as educational reforms (Beijaard et al., 2007; D. L. Butler, Lauscher, Jarvis-Selinger, & Beckingham, 2004). For some of 152

Chapter 6

these assumptions empirical evidence is also available. The study of Kreber, Castleden, Erfani, and Wright (2005), for example, showed that the engagement of university teachers in SRL was related to participation in educational development activities. In addition, it is also often assumed that teachers, who know how to be a self-regulated learner themselves, are also better capable of promoting this kind of learning in their students (Hagger et al., 2008; Niemi, 2002; Van Petegem, Donche, & Vanhoof, 2005). In conclusion, empirical evidence of how SRL makes a difference seems to be a very challenging task, since student teachers need to be followed during their further career and measurements need to be developed for measuring their flexibility in everyday practice, their continuous development, their responses to educational reforms, and the extent to which they promote SRL and in turn contribute to higher learning outcomes in their students.

6.2.3 What makes a difference in self-regulated learning (and why)? In this dissertation, three variables were taken into account to explain differences in SRL. The first variable was the context of learning, the second was the influence of time, and the third was the programme track the student teachers were following. The analysis on the level of the learning experience made clear that regulation of learning at the university is more passive than regulation of learning in practice. The intraclass correlation of the learning orientation scales and regulation dimensions as presented in Chapter 5, showed that for all variables a substantive part of the variance in the data was time-related. Nevertheless, only for six out of twelve variables a growth model could be fitted, from which only for two variables a model with variety in growth rates was accepted. Furthermore, the inclusion of the programme tracks in the models did explain some variance between student teachers. However, a large part of the variance within and between student teachers remained unexplained. In this study we only added the programme track as an explanatory variable. Previous studies have also shown that the perception of the learning environment and personal variables may be related to student teachers’ learning orientations (Oosterheert, Vermunt, & Veenstra, 2002). Besides the unexplained variance between and within student teachers, some of the differences that could be explained were hard to interpret, since they were not in line with our expectations based on the literature. Why do student teachers regulate more passively when learning at the university? What makes student teachers more passive in their regulation over time? What 153

causes the more preferable way of learning of the job-track student teachers? In the previous chapters, we have suggested some possible explanations: in practice there is often no “external regulator” present to rely on, so if a student teacher wants to learn at the practice school at least some active regulation is needed. However, this does not answer the question why at university passive regulation is triggered. Boekaerts and Niemivirta (2000) have suggested that SRL may be hard to accomplish as long as students still have the assumption that the teacher takes responsibility for their learning. Furthermore, they found that students often feel less opportunity and necessity to regulate their learning, than teachers think they have created. Further research is needed to discover if this is also the case in teacher education. Some researchers have expressed concern about the quality of learning in a time-intensive job-track, since the high amount of lessons these student teachers are teaching might provoke an increased focus on the demands from practice and leave not enough time to plan and reflect on learning experiences and integrate theory and practice (Darling-Hammond, 1999; Hagger et al., 2008). Our study has shown that the differences that were found regarding student teachers’ way of learning were in favour of these job-track student teachers. We have seen in our data that the differences between the two tracks in terms of their learning orientations increased, thus the longer they were exposed to the programme, the larger the differences. However, these programmes differed from each other with respect to multiple features and therefore we could not answer the question as to what exactly caused these differences. For example, the student teachers not only taught more lessons in the job-track, but also had more responsibility, were older and spent more time at school. Further research is necessary to study the influence of several contextual and personal factors on student teachers’ self-regulated learning more in detail.

6.3 Towards a better understanding of the nature of student teachers’ self-regulated learning In this dissertation, a concept was studied that is primarily known from studies into academic learning. This means that the theories of SRL are also based on this type of learning. These theories of SRL share some assumptions that do not always fit the assumptions underlying theories of learning to teach. Applying these SRL theories from the context of academic learning to teacher 154

Chapter 6

learning often results in a picture of teachers as non-regulators, in a sense that they often do not plan their learning and are more focused on their teaching behaviour than their own learning process. Below we will discuss the conflicting assumptions based on the results of our studies, to arrive at a better understanding of the characteristics of a self-regulating (student) teacher.

6.3.1 The relevance of planning in self-regulated learning for learning to teach In most theories goal-setting and planning is seen as a necessary condition for SRL: the SRL process starts when goals have been set (Boekaerts & Niemivirta, 2000; Pintrich, 2000; Winne, 1995). SRL theories stress that it is important that students learn not to be distracted by other goals instead of their learning goals, which might increase the time needed to reach these learning goals. Eilam and Aharon (2003, p. 306) state that “without time pressure, there is no regulation needed”. Thus, SRL in traditional student learning theories is aimed at undertaking activities to efficiently reach learning goals. On the other hand, teacher learning is hardly organised around concrete tasks with specific predefined learning goals. Although teacher learning can also be planned, much of teacher learning is unintentional in the sense that it is more ongoing and incidental (Hodkinson & Hodkinson, 2005). Also from the field of workplace learning we know that a large part of learning in informal settings consists of reactive and unconscious learning, and only a small part of learning is deliberative learning where there is “a definite learning goal and time is set aside for acquiring new knowledge” (Eraut, 2004, p. 250). In our small-scale study, 38% of the reported learning experiences were unplanned, in the large-scale study this was the case for 45%. For experienced teachers even two-thirds of the learning experiences were found to be unplanned (Van Eekelen, Boshuizen, & Vermunt, 2005). Nevertheless, deliberate, planned learning to reach goals is also for teachers, and especially for student teachers who have to reach a minimum level of competence, an essential skill to steer their learning in a certain way and not to make learning fully dependent on what happens in practice. The underlying structure in the data showed that this unplanned learning is not the same as unregulated learning. The existence of the retrospective regulation type showed that many learning experiences without pre-set goals in reaction to what happened unexpectedly, were still monitored, evaluated and reflected upon in a very active way. This type of regulation resembles what Iran-Nejad calls dynamic 155

regulation (Iran-Nejad, 1992). Therefore, a theory of SRL in learning to teach focuses both on planned learning processes, but also acknowledges the more reactive regulation processes that are an essential part of learning from experience.

6.3.2 The relevance of “self” in self-regulated learning for learning to teach In research on academic learning, the dimension self-regulation versus external regulation is often used to describe variations in regulation of learning (Kaplan, 2008; Vermunt, 1998). For student teachers and even more for experienced teachers, in most of the learning experiences no teacher educator or manager is present who can or will externally regulate their learning. Nevertheless, an important aspect of adaptive expertise involves the ability to learn from others (Hammerness et al., 2005). Teachers’ learning processes are context-bound and integrated in the community of practice of the school (Bolhuis, 2003). Self-regulation in school learning has been mainly interpreted as an individual activity (Hadwin & Oshige, 2007; Winne, 1995). Traditional theories of SRL often do not deal with the influence of context in which a person learns (Boekaerts, 2002) and do not acknowledge the regulation that is shared or distributed among individuals (Hadwin & Oshige, 2007). In our opinion, self-regulation versus external regulation of learning is for learning to teach a false dichotomy. This easily leads to the misconception that SRL is about whether students have moved along the whole learning process without having used any external support. In a conceptual framework of SRL for learning to teach, SRL is not a question of “who did it”, in a sense that learning should only be considered as self-regulated if the student teacher was the only person who has steered his or her learning process. This would suggest that when someone or something else stimulated their learning, it would not be self-regulated learning anymore. Our study has shown that in learning to teach, the self in SRL refers primarily to the active role that (student) teachers have in their learning. That they are able to direct their learning when necessary, but that they are also able to seek help, search and use external sources to feed their learning, learn in cooperation with others and use their environment for feedback. Learning at the workplace is collaborative in nature (Tynjälä, 2008) since it takes place in communities of practice (Westheimer, 2008). Therefore it is very important on the one hand that teachers are able to let others influence their learning and to take part in this community, but on the other hand to connect it to their own professional 156

Chapter 6

development (D. L. Butler et al., 2004). In this way, the way and amount of influence of the environment and the effect it has on their learning can still be under the control of the learner and be part of their active regulation strategy.

6.3.3 The relevance of “learning” in self-regulated learning for learning to teach Most models of SRL are based on the assumption that goals to direct learning always have to be learning goals (Boekaerts & Corno, 2005; Pintrich, 2000). In informal learning, however, the goals tend to be broader and may also include general behaviour directing goals which may result in more variability in what is learned (Boekaerts & Minnaert, 1999). Our study showed that the activities involved in SRL are often directed towards work-based goals to improve teaching behaviour. Teachers’ goals are more focused on the achievement and well-being of their students than on their personal achievement (R. Butler, 2007). Although teachers focus their regulation activities more on their behaviour, for example on attaining working goals or goals for their students, in our studies this kind of deliberate practice (Ericsson, 2006) was reported as a learning experience by student teachers. From theories of experienced teacher learning we know that significant changes in beliefs and attitudes take place after changes in student learning outcomes are evident (Clarke & Hollingsworth, 2002). Dunn and Shriner (1999) have raised the question whether setting goals for self-improvement would lead to other results than goals for improving instruction and even asked whether this is a worthwhile distinction to make? In our opinion, it is at least as important to teach student teachers how to engage in deliberate practice by creating experiences in which there is a high chance of learning by setting clear work-based goals, as to teach them how to formulate their learning goals. Consequently, in a conceptual framework of SRL for learning to teach, it is too narrow to focus only on setting learning goals and not also on work-based goals as part of SRL.

6.3.4 Conclusion: Portraying the student teacher as a self-regulated learner The conceptual framework of SRL for the context learning to teach that has been developed in our studies showed that student teachers differ on two dimensions in regulating their learning, resulting in three types of regulation: passive regulation, active prospective regulation and active retrospective regulation. The fact that prospective and retrospective regulation defined one of the dimensions showed that this distinction explained differences between 157

learning experiences. In other words, this descriptive framework showed how, in our data, variance in student teachers’ learning experiences could be explained by their differences in focus on prospective or on retrospective regulation. In the previous sections we have shown that unplanned learning is a natural part of learning from practice, but nevertheless active prospective regulation by setting goals and choosing learning strategies is an important aspect to give direction to one’s learning. Therefore, in a prescriptive framework, a high quality of SRL in learning to teach would include both active prospective and active retrospective regulation, although not always necessarily as part of the same learning experience. In our studies we have tried to come to a better understanding of the nature of the dynamic part of student teachers’ regulation of learning in a dual learning programme. We have used existing theories about student teachers’ learning orientations to describe the more stable part of SRL (e.g., Oosterheert & Vermunt, 2001). From these findings regarding the nature of both parts of the concept of SRL for learning to teach, we can conclude student teachers can be seen as selfregulated learners if they: • acknowledge their responsibility to steer their development as a teacher; • are open and able to discuss and reflect on ideas of good teaching and translate this to learning experiences for themselves; • can regulate their learning in an active prospective way by setting learning goals and choosing deliberately appropriate learning strategies to reach them; • can regulate their behaviour in an active prospective way by engaging in deliberate practice to enhance the change to learn from experimenting in the classroom; • can regulate their reactive learning experiences in a retrospective way by connecting them to their ideas of good teaching, reflecting and evaluating them also on the level of the learning process, and reach new forethought on learning; • situate retrospective and passively regulated learning experience in their own process of development leading to new learning experiences; and • use the environment and external sources to feed their learning and in this way to transform the working environment with experiences distracting themselves from their learning process to a learning environment with experiences in which they recognise the learning potential. 158

Chapter 6

6.4 Towards a better understanding of the development of student teachers as self-regulated learners Besides describing the nature of student teachers’ SRL, the second aim of our studies was to increase the understanding of the development of student teachers’ SRL. In this section, we will discuss our findings regarding this development as described in Chapter 5 in relation to the literature and our discussion of the nature of SRL as described above. The results of our study have shown that the dynamic part of SRL as well as student teachers’ learning orientations showed change over time. However, this development was not always in the direction of the description of a self-regulated learner as described in section 6.3.4. In general, student teachers became more passive in their regulation. On the other hand, in terms of retrospectiveness, student teachers’ changed from a focus on prospective regulation to a focus on retrospective regulation, finally ended more in between. This indicates that student teachers have changed from a more onesided to a more versatile focus, which can be seen as one of the characteristics of a self-regulating student teacher. The results regarding changes in learning orientation indicated that student teachers differed in their direction of change. Student teachers changed towards as well as away from independent meaning-oriented learning. We have suggested in Chapter 5 that the increased passiveness of student teachers’ regulation of learning might indicate a period of consolidation (Huberman, 1989), after a period of intense learning. Furthermore, we have raised the possibility that certain barriers to learning (Illeris, 2007) may prevent student teachers further developing towards independent meaning-oriented learning. Several authors have shown the benefit of SRL for learning to teach as well as for being an adaptive expert as a teacher (Hammerness et al., 2005). If we still agree on the view that the expected development of student teachers should be towards a more active and independent meaning-oriented way of learning, it is necessary to get more insights in the factors which are necessary for student teachers to develop as self-regulated learners. In the literature, three conditions have been mentioned that are necessary to change towards more SRL. First, student teachers must feel the will or need to regulate their learning (Boekaerts & Niemivirta, 2000; Van Eekelen, Vermunt, & Boshuizen, 2006). When student teachers are satisfied with how they are performing in practice, or as long as they expect that by 159

working as a teacher their level will improve by itself, they will not feel the need to put energy into regulating their learning in a more active way. Furthermore, if their active regulation turns out not to be worthwhile, it is likely to diminish quickly. Also, questioning a frame of reference is only likely to occur when student teachers have the feeling that their current frame is not sufficient enough. Second, student teachers must have the opportunity to selfregulate their learning (Boekaerts & Niemivirta, 2000; Kremer-Hayon & Tillema, 1999; Vermunt & Verloop, 1999). In general, the dual learning programme seems to give enough opportunities for student teachers to regulate their learning. The opportunity might be limited by time constraints, but also by too strong regulation from others. However, it is worthwhile to find out to what extent, for example, student teachers feel constrained to direct their own learning by the view of the mentor at the practice school on what and how student teachers should learn. Last but not least, even if student teachers feel the need, have the will and the opportunity to become more active, independent and meaningoriented in their learning, they still need to develop the capacity for it (Boekaerts & Corno, 2005; Oosterheert & Vermunt, 2003; Zimmerman & Schunk, 2001). Oosterheert (2001) and Illeris (2007) have shown that sometimes student teachers do not further develop themselves, because there are some emotional barriers to overcome. Furthermore, to direct one’s own learning, it is also important that student teachers are able to think about themselves as learners. Although the citations as mentioned in the General Introduction showed that some student teachers certainly view themselves as learners and acknowledge their responsibilities for their own development, others have more problems with this. For example, seeing oneself as a learner implies that student teachers are not only able to plan what they learn and how this worked out, but also how they learned and how effective this way of learning was. So far, our results have shown that for the majority of the student teachers this is no common practice. In conclusion, our study has identified that student teachers differ in their development as a self-regulated learner. By answering the question what is needed for student teachers to change their way of learning towards more active, independent and meaning-oriented learning, we could also get some indication of what limits this development. Further research is necessary to discover whether and how the three above mentioned conditions stimulate or

160

Chapter 6

hamper student teachers’ development as an active, independent and meaning-oriented learner.

6.5 Reflection on the quality of the studies In this section we will reflect on the quality of the studies which were reported in this dissertation. In these studies we combined different research approaches, developed instruments, collected qualitative and quantitative data and analysed and combined these with data analyses which are not wellknown techniques for everybody. This implies that a lot of decisions had to be made without always having standardised procedures to ensure the quality. The arguments and alternatives that were taken into consideration were often not discussed in detail in the different chapters for sake of readability and because of limited space. In this section we will discuss the quality and limitations of our studies by describing these difficult decisions and our arguments for our decisions, as well as the procedures that were used to ensure the quality of the research project.

6.5.1 Choosing, defining and measuring the concept of self-regulated learning In this dissertation, a concept was studied that is primarily known from studies into academic learning. In the literature, other concepts such as autonomous learning, self-directed learning, independent learning, and active learning have also been used to describe the student’s active role in the learning process (Niemi, 2002). For the context of adult learning, the concept of self-directed learning is often used, which describes how adults pursue learning throughout their lifespan as well as the demand of lifelong learning and learning in life (Bolhuis, 2003; Candy, 1991). Although learning in practice as part of a dual learning programme is closely related to how adults learn from their work, this learning is still embedded in the more formal structure of the teacher education programme. For this more formal context, the models of SRL are more elaborate and studied than those of self-directed learning. Therefore, we chose SRL to be the most relevant concept to describe student teachers’ active involvement in their learning in a dual learning programme. One of the contributions of this dissertation is that we defined SRL as the combination of the more stable learning orientations and the more dynamic regulation activities in concrete learning experiences. In this way we combined 161

different research traditions and different measurement methods to get a more integrated view on student teachers’ self-regulated learning. In the small-scale study, we chose to focus only on student teachers’ learning conceptions to describe the more stable part of SRL, since this concept became closest to the original meaning of the metacognitive knowledge part of SRL as defined by Flavell (1978). In the larger-scale study, we decided to use the concept of learning orientations to study student teachers’ disposition to learning and consequently also included student teachers’ preferences for learning and regulation activities. Initially, we were hesitant about including these aspects, since at first sight they seemed to resemble the conceptualisation of the dynamic part of SRL as measured by the Learning Reports. Closer examination of the ILTP (Oosterheert, Vermunt, & Denessen, 2002), which was used for this study, showed that this questionnaire focused more on general regulation conceptions and role division than on concrete regulation activities. Furthermore, by using the ILTP we were also able to include student teachers’ general emotion regulation, which was not an element of the Structured Learning Report. The psychometric quality of the Structured Learning Report for measuring student teachers’ quality of regulation was already discussed in Chapter 4. The strength of this instrument is that the quality of regulation is defined on the basis of the collection of multiple self-selected learning experiences per student teacher per measurement occasion. On the other hand, in this way the outcomes are also dependent on the selection of these experiences by the student teacher. The number of six learning experiences was chosen to catch variation in regulation of learning within a student teacher. However, we did not collect information about how and why the student teachers selected the specific reported learning experiences. Further research has to show to what extent variations in the amount of learning experiences that have to be reported and the instructions for choosing the experiences influence the outcomes.

6.5.2 Choosing the design of the studies By choosing the concept of SRL that is primarily studied in the context of academic learning, we compelled ourselves to first study the nature of student teachers’ regulation of learning in the specific context of a dual learning programme. Therefore we chose to design our first study as a small-scale and qualitative study that made it possible to study multiple experiences per student teacher to capture the variation in their regulation of learning across 162

Chapter 6

learning experiences. For the second study, many alternatives were considered. One alternative was to study in more detail why student teachers employed certain types of regulation and how this was related to the specific context in which the learning experience occurred. By designing a larger scale study and including more participants we lost this possibility. In our opinion, the most important step to be taken after the small-scale study was to confirm with a larger data set the conceptual framework which we had found. On the other hand, it would have been even better if we could have also included another teacher education institute to increase the generalisability of the results. However, lack of time and supplies made us to skip this ambition. Furthermore, we decided to include both the job-track and the interntrack in the programme. This gave us the opportunity to measure the effect of the track on student teachers’ development in SRL. Unfortunately we were not able to measure differences between the two tracks right from the start of the programme, since the instruments that we used were specifically developed for a dual learning programme in which student teachers learn both at university and from teaching at a practice school. At the start of the programme, the student teachers have no experience yet with this type of learning. Only after three months, we considered student teachers to have enough experience with this type of learning to complete the questionnaires. Therefore, we do not know to what extent existing differences between the student teachers from the two tracks, such as age and prior experience, might also have influenced the results. To ensure that student teachers would not be overloaded with questionnaires and to keep the test-effect as low as possible, we have chosen a three-wave design. If we could have included more measurement occasions, our design would have had more power to test models with variance in slopes in the multilevel analysis. The advantage of this would have been that with models including differences in slopes, we could also have tested the effect of the programme track on the growth rate, instead of only on the intercept value.

6.5.3 Choices during data analyses In our studies several analysis techniques have been combined. To check the categorisation of our qualitative data, an independent researcher was asked to repeat our coding with the developed code book. According to the rule of Cicchetti (1976), 2n2 fragments, in which n is the number of categories, have to be coded by an independent second researcher to gain a reliable interpretation 163

of Cohen’s Kappa. This rule has been applied for estimating the reliability of the coding in the large-scale study (see Chapter 4). The application of this rule to the first study with five to seven categories for eight different variables would have resulted in a minimum of 632 fragments that should be categorised independently by a second independent researcher. After coding of a random selection of 10% (105 fragments) of the total data set a satisfying Kappa was reached except for one variable, for which another 10% was coded after revision of the code book. Since the coding work was very labourintensive, we decided to calculate the Kappa based on this smaller amount of fragments, resulting in an overall Cohen’s Kappa of .90, which can be regarded as reflecting high agreement. However, to make sure that this was not a coincidental easy part of the data, the quality check would have been more reliable with a larger sample. We have used quantitative techniques to make sense of the relation between categories describing the variety of student teachers’ regulation of learning. However, this resulted in a big gap between the richness of more than 50 categories and the structure of two dimensions from the multiple correspondence analyses to describe the variance in our data. For studying individual growth, we had to release the richness even further by using aggregated scores in which multiple learning experiences were also summarised. Therefore, the interpretation of the underlying structure has been crucial for the meaning of the final results. This interpretation was a long and difficult process in which we went back and forth through the data and many alternative interpretations were tested. Unfortunately, in research papers space is limited to describe this process in enough detail to explain the reader the complete process. The results from the multiple correspondence analysis showed how categories are positioned on the dimensions and how the variables contributed to the solution and this formed the main information source for our interpretation. Furthermore, alternative solutions were tested without categories with low frequencies and without outliers. This showed that the solution was stable and even with a selection of the data the interpretation remained the same. To check the face validity of our interpretation, we have discussed the interpretation not only with our own group of researchers, but the results were also presented at 15 conferences, and three other research groups were visited for feedback.

164

Chapter 6

6.6 Implications for practice SRL is an important part of the competence “reflection and development” as defined for teachers in The Netherlands by the Association for the Professional Quality of Teachers (SBL). This association described that a teacher in secondary education “must permanently work on his personal and professional development. That is his/her responsibility, and in order to take this responsibility the teacher must possess competence for reflection and development” (SBL, 2004, p. 20). Although this description gives an idea of the direction in which student teachers should develop, this competence is often mentioned as the most intangible competence by student teachers as well as teacher educators. We hope that the results of our studies have given a clearer picture of what it means to be a self-regulated learner as a student teacher. On the other hand, our study also revealed that giving student teachers the opportunity to direct their own learning (Kremer-Hayon & Tillema, 1999) is not a sufficient condition for most of the student teachers to become an active, independent and meaning-oriented learner. Below we will describe the implications of our results for the current design and pedagogies as often used in dual teacher education programme and give suggestions to further stimulate the development of this competence.

6.6.1 Implications for the design of pre-service teacher education programmes In section 6.4, we have shown that, for increasing student teachers’ SRL, they must feel the need and will to change, and also have the opportunity and the ability. Considering the opportunity, contemporary teacher education programmes are often designed in a way so that student teachers get a lot of freedom in designing their own learning trajectory. Moreover, student teachers also need the opportunity to develop their SRL skills. Vermunt and Verloop (1999) showed that loose external regulation of learning is only effective if the student is capable of at least an intermediate level of self-regulation. They argue that this loose control is only appropriate if students are proficient in particular learning strategies. Although student teachers start the programme while already having obtained a master’s degree, they are beginners in the process of learning to teach in which they have to combine learning from practice and theory. For these student teachers it might be too overwhelming and complicated to comprehend how to learn in such a programme whilst simultaneously being responsible for this learning process. At the end of the 165

programme, student teachers should be able to steer what and how they learn by themselves to be prepared for further learning as an in-service teacher. However, for beginners without having an idea in what direction and how they can regulate learning in a dual learning programme, this task might be too complicated. In particular, student teachers without any experience as a teacher might therefore benefit from some more guidance in the beginning of the programme in what and how they can learn. In Chapter 5, the comparison of two tracks of the programme lead to the conclusion that the student teachers from the more practice-intensive jobtrack of the programme showed more often a meaning-oriented way of learning and scored higher on some of the corresponding learning orientation scales. Our study was not designed in a way that we could answer exactly what in these programmes caused this difference. Therefore, we want to make explicit that our results do not show that a job-track should be preferred for all student teachers over an internship-track. Our study focused only on the reflection and development competence (SBL, 2004) and we did not study the development of the other aspects of teaching. For the development of other competences, the intern-programme might be more effective. For example, Mainhard (2009) suggested that changing schools as interns do, is sometimes the only effective intervention when teachers are struggling with a class, since their reputation at a school hinders a new start. Oosterheert (2001) suggested that changing teaching context might stimulate the development of student teachers’ frame of reference. Our results showed that indeed interns more often changed towards an independent meaning-oriented way of learning. Student teachers with a paid job more often kept the same learning orientation. However, interns also changed more often away from independent meaningoriented learning after changing schools. Therefore, it is important to closely monitor interns’ changes in learning habits after changing schools.

6.6.2 Implications for current pedagogies and tools in pre-service teacher education programmes Since the model of the reflective practitioner is currently one of the most frequently used models for stimulating learning from experience in teacher education programmes in The Netherlands, we would like to connect our findings to this perspective. It might be confusing that the concept of reflection is also used in SRL theories to describe the learning phase in which students look back on their learning to describe what they have learned and how their 166

Chapter 6

way of learning contributed to this (Vermunt & Van Rijswijk, 1988; Zimmerman, 2000). One of the main criticisms on the concept of reflection is also that its widespread use has lead to a lack of consensus of the meaning of the concept and how it should be applied (Mena Marcos, Sánchez Miguel, & Tillema, 2009). Nevertheless, several studies have shown the benefits of reflective skills for learning to teach (e.g., Korthagen & Kessels, 1999). In Dutch pre-service teacher education programmes, often the ALACT model of reflection is used (Korthagen, 2005). This is a circular model with five subsequent phases starting with (1) Action, followed by (2) Looking back on the action, (3) Awareness of essential aspects, (4) Creating alternative methods of action and (5) Trial, which can be the start of a new (1) Action. This model shows how student teachers can learn from their action, but also that reflective learning relies, to a great extent, on the action that already has taken place. This model gives less guidance to regulate learning based on inspiration from theory or from their own frame of reference, and how to engage in deliberate practice. Also the model of Clarke and Hollingsworth shows that besides reflection, enactment is also one of the mediating processes in experienced teacher learning (Clarke & Hollingsworth, 2002). Reflection therefore fits very well in a retrospective way of regulation, but does not necessarily stimulate active prospective regulation. Conway (2001) therefore advocated more focus on anticipatory reflection to broaden the term from exclusively looking back in time. In conclusion, although reflection is a key element in learning to teach from practice, an exclusive focus on this element will be too limited to help student teachers to direct their learning. A frequently used tool for stimulating SRL is the portfolio (e.g., Kicken, Brand-Gruwel, Van Merriënboer, & Slot, 2009; Meeus, Van Petegem, & Meijer, 2008; Strijbos, Meeus, & Libotton, 2007). By asking student teachers to make personal development plans, reflect on them and keep logs of their experiences, it is often assumed that their SRL will increase. The study of Mansvelder-Longayroux (2006), however, showed that the learning portfolio in itself is not well suited to stimulate learning activities such as analysis, critical processing, diagnosis, and reflection. It is too easy to assume that regulation of the learning process will follow from the production of a portfolio (Mansvelder-Longayroux, 2006; Oosterheert, 2001): “Teacher education institutes should pay attention to the extent to which they appeal to student teachers’ capacity for SRL by using portfolio and, if necessary, they must help and supervise the student teachers more in the process of learning 167

from experiences” (Mansvelder-Longayroux, 2006, p. 110). One of the reasons that the portfolio might have limited effects on student teachers’ SRL is that regulation of learning to teach is a more ongoing process in which the classroom experiences are far more influential to their learning than the professional development plans they have written some weeks ago. The process of regulation of learning can take place on different levels, from which reflecting on their general development and formulating new development goals, is only one part of it. However, how to regulate concrete learning experiences situated in the complex classroom practice is a completely different skill. Some student teachers might be very well capable of doing this, but if we want to stimulate all student teachers’ development in SRL, more concrete assignments are needed to teach student teachers how to regulate their learning. Some suggestions for pedagogies and tools to stimulate SRL will be given in the next section.

6.6.3 Suggestions and tools for stimulating self-regulated learning A starting point for stimulating the will and ability to change is to make student teachers aware of how they learn. Both the ILTP and the Structured Learning Report can be used as tools for this purpose. As a reward for their participation in our large-scale study, student teachers received a personal report of how they learnt and how this changed throughout the programme. As described in Chapter 4, student teachers responded positively to using the instruments: two-thirds of the respondents spontaneously described that they had liked participating in this study, and how much they had learned from participating and reading their personal description of their development as a learner. Only two student teachers answered that filling out the instruments (throughout the year a total number of 21 questionnaires), asked too much of their time. Although most of the student teachers were positive, it might happen that student teachers do not agree with the results from the instruments. In these cases the outcomes of the instruments can also be used to discuss in a more concrete way the student teachers’ conceptions of learning. A good starting point for this discussion might be to talk about what kind of learning they expect from their own students and how they try to stimulate this as a teacher. This might stimulate their thinking about differences in learning, which might help them to recognise these differences also in their own and other student teachers’ learning.

168

Chapter 6

A second step in stimulating the will and ability of student teachers to develop their way of learning is to give them an overview of alternative ways of learning. If student teachers are responsible for their own learning and for the choices they make, they also need to know what their options are. For example, in the personal report as described above, student teachers also received some theoretical background about individual differences in learning to teach and a description of the variety in regulation activities was included. Student teachers might experience their own way of learning as the most logical or sometimes even only possible way of learning. It might also be helpful for expanding their repertoire to discuss learning experiences with a peer who has a totally different way of learning. The student teachers could explain to each other what kind of learning and regulation strategies they use, why and why this works for them. For example, one student teacher from the small-scale study responded to the question: Did you plan to learn this?, with the reaction that she had no idea how it would be possible to plan your learning: “If you plan your learning, you already know what you are going to learn and then learning already has taken place”. For this student teacher it might be helpful to hear from peers how they are planning their learning. Also the frequencies of the answers on the question about their decision for a learning strategy showed that only a very small section of student teachers ever made a deliberate choice for a certain learning strategy (see Chapter 2 and 4).This also showed that student teachers are often not aware of the variety of ways of learning to teach. The suggestions as described above can be seen as general suggestions for teacher educators to incorporate and experiment with in their teaching. Besides, this study has also shown that student teachers vary in how they regulate their learning, in their learning orientations and in their development in these. Furthermore, they differ in the programme track they are following, the specific context of their teaching practice, and their different backgrounds which may also influence their development. Therefore, also student teacherspecific strategies will be needed to support them in their development as an active, independent and meaning-oriented learner, which will be dependent on what kind of change the student teacher has to make and his or her specific context and background. Oosterheert (2001) has suggested how student teachers with a specific learning orientation can be supported in developing their way of learning. For some student teachers, certain barriers to learning (Illeris, 2007) may first need to be identified, before they can develop further. 169

Other student teachers might need a more concrete approach, for example, suggestions for experimenting with other learning and regulation activities. Therefore, the teacher educator has also an important role in identifying student teachers’ individual needs and choosing the specific strategies to support student teachers with their development as a self-regulated learner.

6.7 Suggestions for further research Besides giving answers, this thesis has also raised many new questions. As already suggested in section 6.2 (Questions without answers), the relation between the two parts of SRL and the benefits of SRL are some of the topics that need more research. In previous studies, also often no relation has been found between the general learning orientations and regulation activities in concrete learning experiences. Therefore, a different approach to studying these relations might be needed. For example, by also measuring the conditions under which certain combinations of learning orientations and types of regulation occur, more insight can be obtained into why student teachers show certain combinations. To understand why student teachers learn in a specific way, further research is necessary considering the personal and contextual factors influencing their learning orientations and regulation patterns. In particular, since student teachers learn in different practice schools with even more variance in their supervisors, it would be relevant to find out how this practice context influences the way student teachers learn. This study was carried out in a one-year teacher education programme preparing for upper secondary education. To get more insights in how student teachers develop as self-regulated learners, it is necessary to follow them over a longer period of time. Since other teacher education programmes for lower secondary education and primary education in The Netherlands take four years, it would be interesting to set up a longitudinal study in these contexts that may last longer than one year. In this way, the measurement occasions can increase and more power can be obtained to study different growth rates and also how these are influenced by contextual factors. Moreover, following student teachers after their graduation to see how their way of learning changes after entering the profession, could give more insights into the relevance of SRL for becoming an expert teacher. However, then also different measurement instruments need to be developed for this measuring SRL in the in-service context. 170

Chapter 6

Last but not least, in the past decade, three dissertations in The Netherlands have shown that more specific pedagogies need to be developed to enhance different aspects of student teachers’ learning to teach: reflective skills (Mansvelder-Longayroux, 2006), meaning-oriented learning (Oosterheert, 2001), and self-regulated learning (this dissertation). All of these studies were descriptive in nature and concluded with specific suggestions for improving practice. Nevertheless, enacting ideas in practice is seen in literature as one of the most challenging tasks in teaching. Therefore, the issue of how teacher educators can be supported in translating ideas from research into practice is also an important research topic in itself that must not be forgotten.

171

san diego

higher orchestration psychologist hillsdale

azevedo

self-regulated butler

student teachers'

erlbaum entwistle relationships

component

mutton

theory

alexander

achievement

journal

routledge techniques

beijaard

boulton-lewis

environments

analysis

nonlinear

minnaert

publications

behavior

london

associates quality

evaluation

applied

principal

measuring

preparation

simons

active

model

become

promote

reflection

self-directed dordrecht

differences growth

assessment

statistics

publishers

conceptual

mentors

teacher presented

hagger

education

corno

investigating

lonka

feiman-nemser classroom

metacognition theoretical

cascallar

consequences

stat

methodological

academic

vermunt san fransisco

patterns styles

oosterheert donche

van hout

performance

qualitative

retrieved

psychology

changing

mahwah marton

learning

consistency

examining

motivation ucla relations conceptions

british aspects dissonance perry verloop preservice

demers

winne

annual

boekaerts

participants

situated

utrecht

sage

understanding

wolters

berliner darling-hammond

the netherlands

exploratory

personal

new york

meijer

issues

preprints

mathematics

international

hammerness

data

multilevel

collaborative

lifelong

shared

korthagen

co-regulation

factors

context

review

contemporary secondary

bolhuis

using oxford

epistemological

research

beliefs

american

bergen plenum

social professional expert regulation vermetten students improve mcintyre lawrence approaches knowledge european conference chicago

development

goals introduction individual learners

pintrich primary

effective independent

flavell

university

college aera

department

workplace

perspective

handbook

cognitive

van petegem role

tillema

meeus

skills identity schunk

zimmerman dam

portfolio meetingstrategies schools

informal

cochran-smith

longitudinal

validity

leeuw

teaching orientations

reliability

instruction

outcomes

zeidner

meyer edu

bransford

pre-service pedagogy

References Alexander, P. A. (2008). Why this and why now? Introduction to the special issue on metacognition, self-regulation, and self-regulated learning. Educational Psychology Review, 20, 369-372. Azevedo, R. (2009). Theoretical, conceptual, methodological, and instructional issues in research on metacognition and self-regulated learning: A discussion. Metacognition and Learning, 4, 87-95. Bakkenes, I., Vermunt, J. D., & Wubbels, T. (in press). Teacher learning in the context of educational innovation: Learning activities and learning outcomes of experienced teachers. Learning and Instruction. Beijaard, D., Korthagen, F., & Verloop, N. (2007). Understanding how teachers learn as a prerequisite for promoting teacher learning. Teachers and Teaching: theory and practice, 13, 105-108. Beijaard, D., Meijer, P. C., & Verloop, N. (2004). Reconsidering research on teachers’ professional identity. Teaching and Teacher Education, 20, 107-128. Beishuizen, J., Stoutjesdijk, E., & Van Putten, K. (1994). Studying textbooks: Effects of learning styles, study task, and instruction. Learning and Instruction, 4, 151-174. Bembenutty, H. (2007, April). Preservice teachers’ motivational beliefs and self-regulation of learning. Paper presented at the Annual meeting of the AERA, Chicago. Berliner, D. C. (1988). The development of expertise in pedagogy. Washington, DC: AACTE Publications. Berliner, D. C. (2000). A personal response to those who bash teacher education. Journal of Teacher Education, 51, 358. Berliner, D. C. (2001). Learning about and learning from expert teachers. International Journal of Educational Research, 35, 463-482. Biggs, J. (1979). Individual differences in study processes and the quality of the learning outcomes. Higher Education, 8, 381-394. Biggs, J. (1988). Approaches to learning and essay writing. In R. R. Schmeck (Ed.), Learning strategies and learning styles (pp. 185-228). New York: Plenum Press. Boekaerts, M. (2002). Bringing about change in the classroom: strengths and weaknesses of the self-regulated learning approach - EARLI presidential address, 2001. Learning and Instruction, 12, 589-604. Boekaerts, M., & Cascallar, E. (2006). How far have we moved toward the integration of theory and practice in self-regulation? Educational Psychology Review, 18, 199210. Boekaerts, M., & Corno, L. (2005). Self-regulation in the classroom: A perspective on assessment and intervention. Applied Psychology: An International Review, 54, 199-231. 173

Boekaerts, M., & Minnaert, A. (1999). Self-regulation with respect to informal learning. International Journal of Educational Research, 31, 533-544. Boekaerts, M., & Niemivirta, M. (2000). Self-regulated learning: Finding a balance between learning goals and ego-protective goals. In M. Boekaerts, P. R. Pintrich & M. Zeidner (Eds.), Handbook of self-regulation (pp. 417-450). San Diego: Academic Press. Bolhuis, S. (2003). Towards process-oriented teaching for self-directed lifelong learning: a multidimensional perspective. Teaching and Teacher Education, 13, 327-347. Bolhuis, S., & Voeten, M. J. M. (2001). Toward self-directed learning in secondary schools: what do teachers do? Teaching and Teacher Education, 17, 837-855. Boulton-Lewis, G. M., Wilss, L., & Lewis, D. (2003). Dissonance between conceptions of learning and ways of learning for indigenous Australian university students. Studies in Higher Education, 28, 79-90. Boyle, M. H., & Willms, J. D. (2008). Multilevel modelling of hierarchical data in developmental studies. Journal of Child Psychology and Psychiatry, 42, 141-162. Brown, A. L. (1978). Knowing when, where, and how to remember: A problem of metacognition. In R. Glaser (Ed.), Advances in instructional psychology (pp. 77165). Hillsdale: Erlbaum. Brown, A. L. (1987). Metacognition, executive control, self-regulation, and other more mysterious mechanisms. In F. E. Weinert & R. H. Kluwe (Eds.), Metacognition, Motivation, and Understanding (pp. 65-116). Hillsdale: Erlbaum. Brown, A. L., & DeLoache, J. S. (1978). Skills, plans, and self-regulation. In R. S. Siegler (Ed.), Children's thinking: what develops? (pp. 3-37). Hillsdale: Erlbaum. Brownlee, A., Purdie, N., & Boulton-Lewis, G. M. (2001). Changing epistemological beliefs in pre-service teacher education. Teaching in Higher Education, 6, 247268. Brownlee, A., Purdie, N., & Boulton-Lewis, G. M. (2003). An investigation of student teachers’ knowledge about their own learning. Higher Education, 45, 109-125. Buehl, M. M., & Alexander, P. A. (2006). Examining the dual nature of epistemological beliefs. International Journal of Educational Research, 45, 28-42. Buitink, J. (2009). What and how do student teachers learn during school-based teacher education. Teaching and Teacher Education, 25, 118-127. Butler, D. L. (2002). Qualitative approaches to investigating self-regulated learning: contributions and challenges. Educational Psychologist, 37, 59-63. Butler, D. L., Lauscher, H. N., Jarvis-Selinger, S., & Beckingham, B. (2004). Collaboration and self-regulation in teachers’ professional development. Teaching and Teacher Education, 20, 435-455. Butler, R. (2007). Teachers’ achievement goal orientations and associations with teachers’ help seeking: Examination of a novel approach to teacher motivation. Journal of Educational Psychology, 99, 241-252. 174

References

Candy, P. C. (1991). Self-direction for lifelong learning. San Fransisco: Jossey-Bass. Cano, F. (2005). Consonance and dissonance in students’ learning experience. Learning and instruction, 15, 201-224. Cantwell, R. H., & Moore, P. J. (1996). The development of measures of individual differences in self-regulatory control and their relationship to academic performance. Contemporary Educational Psychology, 21, 500-517. Cascallar, E., Boekaerts, M., & Costigan, T. (2006). Assessment in the evaluation of selfregulation as a process. Educational Psychology Review, 18, 297-306. Cattell, R. B. (1966). The scree test for the number of factors. Multivariate Behavioral Research, 1, 245-276. Chen, P. P., & Bembenutty, H. (2005). Self-efficacy of urban preservice teachers. Academic Exchange Quarterly, 9(4), 273-280. Cicchetti, D. V. (1976). Assessing inter-rater reliability for rating scales: Resolving some basic issues. British Journal of Psychiatry, 129, 452-456. Clarke, D., & Hollingsworth, H. (2002). Elaborating a model of teacher professional growth. Teaching and Teacher Education, 18, 947-967. Cochran-Smith, M., Feiman-Nemser, S., McIntyre, D. J., & Demers, K. E. (Eds.). (2008). Handbook of research on teacher education. Enduring questions in changing contexts (3rd ed.). New York: Routledge. Cochran-Smith, M., & Zeichner, K. (Eds.). (2005). Studying teacher education: The report of the AERA panel on research and teacher education. Mahwah, New Jersey: Lawrence Erlbaum Associates, Publishers. Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159. Conway, P. F. (2001). Anticipatory reflection while learning to teach: from a temporally truncated to a temporally distributed model of reflection in teacher education. Teaching and Teacher Education, 17, 89-106. Conway, P. F., & Clark, C. M. (2003). The journey inward and outward: a reexamination of Fuller's concerns-based model of teacher development. Teaching and Teacher Education, 19, 465-482. Cornford, I. A. (2002). Learning-to-learn strategies as a basis for effective lifelong learning. International Journal of Lifelong Education, 21, 357-368. Corno, L. (1986). The metacognitive control components of self-regulated learning. Contemporary Educational Psychology, 11, 333-346. Corno, L., & Mandinach, E. B. (1983). The role of cognitive engagement in classroom learning and motivation. Educational Psychologist, 18(2), 88-108. Corrigan, G., & Taylor, N. (2004). An exploratory study of the effect a self-regulated learning environment has on pre-service primary teachers’ perceptions of teaching science and technology. International Journal of Science and Mathematics Education, 2, 45-62.

175

Darling-Hammond, L. (1999). The case for university-based teacher education. In R. Roth (Ed.), The role of university in the preparation of teachers (pp. 13-30). New York: Routledge/Falmer. Darling-Hammond, L., Hammerness, K., Grossman, P., Rust, F., & Shulman, L. (2005). The design of teacher education programmes. In L. Darling-Hammond & J. Bransford (Eds.), Preparing teachers for a changing world: what teachers should learn and be able to do (pp. 390-441). San Fransisco: Josey-Bass. Darling-Hammond, L., & Snyder. (2000). Authentic assessment of teaching in context. Teaching and Teacher Education, 16, 523-545. De Leeuw, J. (1982). Nonlinear principal Component Analysis. Paper presented at the COMPSTAT, Toulouse. De Leeuw, J. (2005a). Nonlinear principal component analysis. Department of Statistics Papers. Retrieved 03-25, 2005, from http://respositories.cdlib.org/uclastat/ papers/2005070801 De Leeuw, J. (2005b). Nonlinear principal component analysis and related techniques [Electronic Version]. Department of Statistics Papers. Retrieved 2009-03-25 from http://preprints.stat.ucla.edu/. Dignath, C., & Büttner, G. (2008). Components of fostering self-regulated learning among students. A meta-analysis on intervention studies at primary and secondary school level. Metacognition and learning, 3, 231-264. Dinsmore, D. L., Alexander, P. A., & Loughlin, S. M. (2008). Focusing the conceptual lens on metacognition, self-regulation, and self-regulated learning. Educational Psychology Review, 20, 391-409. Donche, V., & Van Petegem, P. (2005). Assessing preservice teachers’ orientations to learning to teach and preferences for learning environments. Scientia Paedagogica Experimentalis, 17, 27-52. Donche, V., & Van Petegem, P. (2009). The development of learning patterns of student teachers: a cross-sectional and longitudinal study. Higher Education, 57, 463475. Donche, V., Vanhoof, J., & Van Petegem, P. (2003, April). Beliefs about learning environments: How do student teachers think, reflect and act concerning self-regulated and cooperative learning in Flanders (Belgium). Paper presented at the Annual meeting of the AERA, Seattle. Dunn, T. G., & Shriner, C. (1999). Deliberate practice in teaching: what teachers do for self-improvement. Teaching and Teacher Education, 15, 631-651. Edwards, A., & Mutton, T. (2007). Looking forward: rethinking professional learning through partnership arrangements in Initial Teacher Education. Oxford Review of Education, 33, 503-519. Eilam, B., & Aharon, I. (2003). Students’ planning in the process of self-regulated learning. Contemporary Educational Psychology, 28, 304-334.

176

References

Entwistle, N. (1988). Approaches to learning and essay writing. In R. R. Schmeck (Ed.), Motivational factors in students’ approaches to learning (pp. 21-51). New York: Plenum Press. Entwistle, N., & McCune, V. (2004). The conceptual bases of study strategy inventories. Educational Psychology Review, 16, 325-345. Entwistle, N., & Ramsden, P. (1983). Understanding student learning. London: Croom Helm. Eraut, M. (2004). Informal learning in the workplace. Studies in Continuing Education, 26, 247-273. Ericsson, K. A. (2006). The influence of experience and deliberate practice on the development of superior expert performance. In K. A. Ericsson, N. Charness, P. J. Feltovich & R. R. Hoffman (Eds.), The Cambridge Handbook of Expertise and Expert Performance (pp. 683-703). New York: Cambridge University Press. Evensen, D. H., Salisbury-Glennon, J. D., & Glenn, J. (2001). A qualitative study of six medical students in problem-based curriculum: Toward a situated model of self-regulation. Journal of Educational Psychology, 93, 659-676. Fallon, D. (2008). Perspectives on teacher learning. Normative, logical, empirical. In M. Cochran-Smith, S. Feiman-Nemser, D. J. McIntyre & K. E. Demers (Eds.), Handbook of research on teacher education (pp. 835-841). New York: Routledge. Feiman-Nemser, S. (2001). From preparation to practice: Designing a continuum to strengthen and sustain teaching. Teachers College Record, 103, 1013-1055. Feiman-Nemser, S. (2008). Teacher learning. How do teachers learn to teach. In M. Cochran-Smith, S. Feiman-Nemser, D. J. McIntyre & K. E. Demers (Eds.), Handbook of research on teacher education (pp. 697-705). New York: Routledge. Flavell, J. H. (1971). First discussant's comments: What is memory development the development of? Human Development, 14, 272-278. Flavell, J. H. (1976). Metacognitive aspects of problem solving. In L. B. Resnick (Ed.), The nature of intelligence (pp. 231-235). Hillsdale: Erlbaum. Flavell, J. H. (1978). Metacognitive development. In J. M. Scandura & C. J. Brainerd (Eds.), Structural Process Models of Complex Human Behavior (pp. 213-245). Alphen a.d. Rijn, Netherlands: Slijthoff & Noordhoff. Fuller, F. F., & Parsons, J. S. (1972). Current research on the concerns of teachers. Paper presented at the Annual meeting of the American Educational Research Association. Funnell, B., Bryer, F., & Grimbeek, P. (2004, December). Demographic profiling for educational researchers: Using Optimal Scaling to identify distinct groups of participants. Paper presented at the 2nd Annual International Conference on Cognition, Language and Special Education Research, Gold Coast. Greene, J. A., & Azevedo, R. (2009). A macro-level analysis of SRL processes and their relations to the acquisition of a sophisticated mental model of a complex system. Contemporary Educational Psychology, 34, 18-29. 177

Hadwin, A. F., & Oshige, M. (2007, August). Self-regulation, co-regulation, and socially shared regulation: Examining the many faces of social in models of SRL. Paper presented at the 12th Biennial Conference for Research on Learning and Instruction (EARLI), Budapest. Hager, P. (2004). Conceptions of learning and understanding at work. Studies in Continuing Education, 26, 3-17. Hagger, H., Burn, K., Mutton, T., & Brindley, S. (2008). Practice makes perfect? Learning to learn as a teacher. Oxford Review of Education, 34, 159-178. Hagger, H., & McIntyre, D. (2000). What can research tell us about teacher education? Oxford Review of Education, 26, 483-494. Hammerness, K., Darling-Hammond, L., Bransford, J., Berliner, D. C., Cochran-Smith, M., McDonald, M., Zeichner, K. (2005). How teachers learn and develop. In L. Darling-Hammond & J. Bransford (Eds.), Preparing teachers for a changing world: what teachers should learn and be able to do (pp. 359-389). San Fransisco: JoseyBass. Hedeker, D. (2004). An introduction to growth modeling. In D. Kaplan (Ed.), Quantitative methodology for the social sciences (pp. 215-234). Thousand Oaks CA: Sage Publications. Hodkinson, H., & Hodkinson, P. (2005). Improving schoolteachers’ workplace learning. Research Papers in Education, 20, 109-131. Hoekstra, A., Brekelmans, M., Beijaard, D., & Korthagen, F. (2009). Experienced teachers’ informal learning: Learning activities and changes in behavior and cognition. Teaching and Teacher Education, 25, 663-673. Howard-Rose, D., & Winne, P. H. (1993). Measuring component and sets of cognitive processes in self-regulated learning. Journal of Educational Psychology, 85, 591604. Hox, J. J. (2000). Multilevel analysis of grouped and longitudinal data. In T. D. Little, K. U. Schnabel & J. Baumert (Eds.), Modeling longitudinal and multilevel data: Practical issues, applied approaches, and specific examples (pp. 15-32). Mahwah, NJ: Lawrence Erlbaum Associates. Hox, J. J. (2002). Multilevel Analysis. Techniqus and applications. Mahwah, NJ: Lawrence Erlbaum Associates. Huberman, M. (1989). On teachers’ careers: Once over lightly, with a broad brush. International Journal of Educational Development, 13, 347-362. Hwang, Y. S., & Vrongistinos, K. (2002). Elementary in-service teachers’ self-regulated learning strategies related to their academic achievements. Journal of Instructional Psychology, 29, 147-154. Illeris, K. (2007). How we learn. Learning and non-learning in school and beyond. London: Routledge. Iran-Nejad, A. (1992). Contributions of active and dynamic self-regulation to learning. Innovative Higher Education, 17, 125-136. 178

References

Järvenoja, H., & Järvelä, S. (2007, August). Do students regulate their emotions together in collaborative learning situations? Paper presented at the 12th Biennial Conference for Research on Learning and Instruction (EARLI), Budapest. Jonsson, A., & Svingby, G. (2007). The use of scoring rubrics: Reliability, validity and educational consequences. Educational Research Review, 2, 130-144. Kagan, D. M. (1992). Professional growth among preservice and beginning teachers. Review of Educational Research, 62, 129-169. Kaplan, A. (2008). Clarifying metacognition, self-regulation, and self-regulated learning. Educational Psychology Review, 20, 477-484. Kicken, W., Brand-Gruwel, S., Van Merriënboer, J. J. G., & Slot, W. (2009). Design and evaluation of a development portfolio: how to improve students’ self-directed learning skills. Instructional Science, 37, 453-473. Korthagen, F. A. J. (2005). Practice, theory, and person in life-long learning. In D. Beijaard, P. C. Meijer, G. Morine-Dershimer & H. H. Tillema (Eds.), Teacher professional development in changing conditions. Dordrecht, The Netherlands: Springer. Korthagen, F. A. J. (in press). Situated learning theory and the pedagogy of teacher education: Towards an integrative view of teacher behavior and teacher learning. Teaching and Teacher Education. Korthagen, F. A. J., & Kessels, J. P. A. M. (1999). Linking theory and practice: Changing the pedagogy of teacher education. Educational Researcher, 28(4), 4-17. Kramarski, B., & Michalsky, T. (2009). Investigating Preservice Teachers’ Professional Growth in Self-Regulated Learning Environments. Journal of Educational Psychology, 101, 161-175. Kreber, C., Castleden, H., Erfani, N., & Wright, T. (2005). Self-regulated learning about university teaching: an exploratory study. Teaching in Higher Education, 10, 7597. Kremer-Hayon, L., & Tillema, H. H. (1999). Self-regulated learning in the context of teacher education. Teaching and Teacher Education, 15, 507-522. Lin, X., Schwartz, D. L., & Hatano, G. (2005). Toward teachers’ adaptive metacognition. Educational Psychologist, 40, 245-255. Lindblom-Ylänne, S. (2003). Broadening an understanding of the phenomenon of dissonance. Studies in Higher Education, 28, 63-78. Lindblom-Ylänne, S., & Lonka, K. (1999). Individual ways of interacting with the learning environment — are they related to study success? Learning and Instruction, 9, 1-18. Lindblom-Ylänne, S., & Lonka, K. (2000). Dissonant study orchestrations of highachieving university students. European Journal of Psychology of Education, 15, 19-32. Lonka, K., & Lindblom-Ylänne, S. (1996). Epistemologies, conceptions of learning, and study practices in medicine and psychology. Higher Education, 31, 5-24. 179

Lonka, K., Olkinuora, E., & Mäkinen, J. (2004). Aspects and prospects of measuring studying and learning in higher education. Educational Psychology Review, 16, 301-323. Lortie, D. C. (1975). Schoolteacher: a sociological study. Chicago: University of Chicago Press. Maandag, D. W., Deinum, J. F., Hofman, W. H. A., & Buitink, J. (2007). Teacher education in schools: an international comparison. European Journal of Teacher Education, 30, 151-173. Mainhard, M. T. (2009). Time consistency in teacher-class relationships. Unpublished doctoral dissertation. Utrecht University. Mair, P., & De Leeuw, J. (2008). Rank and set restrictions for homogeneity analysis in R: The "homals" package [Electronic Version]. UCLA Statistics Preprint Series. Retrieved 2009-03-26 from http://preprints.stat.ucla.edu/. Makinen, J., & Olkinuora, E. (2004). University students’ situational reaction tendencies: reflections on general study orientations, learning strategies, and study success. Scandinavian Journal of Educational Research, 48, 477-491. Mansvelder-Longayroux, D. (2006). The learning portfolio as a tool for stimulating reflection by student teachers. Leiden: Drukker Mostert & Van Onderen! Marsh, H. W. (2007). Do university teachers become more effective with experience? A multilevel growth model of students’ evaluations of teaching over 13 years. Journal of Educational Psychology, 99, 775-790. Marton, F., Dall'Alba, G., & Beaty, E. (1993). Conceptions of learning. International Journal of Educational Research, 46, 277-300. Marton, F., & Säljö, R. (1976). On qualitative differences in learning: I - Outcome and process. British Journal of Educational Psychology, 46, 4-11. McIntyre, D., & Hagger, H. (1992). Professional development through the oxford internship model. British Journal of Educational Studies, 40, 264-283. Meeus, W., Van Petegem, P., & Engels, N. (2009). Validity and reliability of portfolio assessment in pre-service teacher education. Assessment & Evaluation in Higher Education, 34, 401-413. Meeus, W., Van Petegem, P., & Meijer, J. (2008). Stimulating independent learning. Educational Studies, 34, 469-482. Meirink, J., Meijer, P. C., Verloop, N., & Bergen, T. C. M. (2009). Understanding teacher learning in secondary education: The relations of teacher activities to changed beliefs about teaching and learning. Teaching and Teacher Education, 25, 89-100. Mena Marcos, J. J., Sánchez Miguel, E., & Tillema, H. H. (2009). Teacher reflection on action: what is said (in research) and what is done (in teaching). Reflective Practice, 10, 191-204. Mevarech, Z., & Kramarski, B. (1997). IMPROVE: a multidimensional method for teaching mathematics in heterogeneous classrooms. American Educational Research Journal, 34, 365-394. 180

References

Meyer, J. H. F. (1991). Study orchestration: the manifestation, interpretation and consequences of contextualised approaches to studying. Higher Education, 22, 297. Meyer, J. H. F. (1999). On the operationalisation of conceptions of learning in higher education and their association with students’ knowledge and experiences of their learning. Higher Education Research & Development, 19, 289-302. Meyer, J. H. F. (2000). The modelling of dissonant study orchestration in higher education. European Journal of Psychology of Education, 15, 5-18. Muhr, T. (1997). ATLAS.ti : the knowledge workbench: visual qualitative data, analysis, management, model building : short user's manual. Berlin: Scientific Software Development. Muthén, L. K., & Muthén, B. (2002). How to use a Monte Carlo study to decide on sample size and determine power. Structural Equation Modeling, 4, 599-620. Mutton, T., Burn, K., & Hagger, H. (2010). Making sense of learning to teach: learners in context. Research Papers in Education, 25, 1-18. Niemi, H. (2002). Active learning - a cultural change needed in teacher education and schools. Teaching and Teacher Education, 19, 763-780. Oosterheert, I. E. (2001). How student teachers learn. Groningen: UCLO. Oosterheert, I. E., Vermunt, J., & Veenstra, R. (2002). Manieren van leren onderwijzen en relaties met persoonsgebonden en contextuele variabelen [Orientations towards learning to teach and relations to personal and contextual variables]. Pedagogische Studiën, 79, 251-268. Oosterheert, I. E., & Vermunt, J. D. (2001). Individual differences in learning to teach: relating cognition, regulation and affect. Learning and Instruction, 11, 133-156. Oosterheert, I. E., & Vermunt, J. D. (2003). Knowledge construction in learning to teach: the role of dynamic sources. Teachers and Teaching: theory and practice, 9, 157-173. Oosterheert, I. E., Vermunt, J. D., & Denessen, E. (2002). Assessing orientations to learning to teach. British Journal of Educational Psychology, 72, 41-64. Orange, C. (1999). Using peer modeling to teach self-regulation. The Journal of Experimental Education, 68, 21-39. Perry, N. (2008). Talking about teaching self-regulated learning: Scaffolding student teachers’ development and use of practices that promote self-regulated learning. International Journal of Educational Research, 47, 97-108. Perry, N., Phillips, L., & Dowler, J. (2004). Examining features of tasks and their potential to promote self-regulated learning. Teachers College Record, 106, 18541878. Perry, N. E. (2002). Introduction: Using qualitative methods to enrich understandings of self-regulated learning. Educational Psychologist, 37, 1-4. Perry, N. E., & Winne, P. H. (2006). Learning from Learning Kits. Educational Psychology Review, 18, 211-228. 181

Peugh, J. L., & Enders, C. K. (2005). Using the SPSS Mixed Procedure to fit crosssectional and longitudinal multilevel models. Educational and Psychological Measurement, 65, 717-741. Pieschl, S., Bromme, R., & Porsch, T. (2008). Epistemological sensitisation causes deeper elaboration during self-regulated learning. Paper presented at the ICLS, Utrecht, The Netherlands. Pintrich, P. R. (2000). The role of goal orientation in self-regulated learning. In M. Boekaerts, P. R. Pintrich & M. Zeidner (Eds.), Handbook of self-regulation (pp. 451-502). San Diego: Academic Press. Pintrich, P. R. (2004). A conceptual framework for assessing motivation and selfregulated learning in college students. Educational Psychology Review, 16, 385407. Pintrich, P. R., & De Groot, E. V. (1990). Motivational and self-regulated learning components of classroom academic performance. Journal of Educational Psychology, 82, 33-40. Pintrich, P. R., & Smith, D. A. F. (1993). Reliability and predictive validity of the motivated strategies for learning questionnaire (MSLQ). Educational and Psychological Measurement, 53, 801-813. Randi, J. (2004). Teachers as self-regulated learners. Teachers College Record, 106, 18251853. Raudenbush, S. W., & Bryk, A. S. (2002). Hierarchical linear models (2nd ed.). Thousand Oaks, CA: Sage Publications. Rodgers, C. R., & Scott, K. H. (2008). The development of the personal self and professional identity in learning to teach. In M. Cochran-Smith, S. FeimanNemser, D. J. McIntyre & K. E. Demers (Eds.), Handbook of research on teacher education (pp. 732-755). New York: Routledge. Ross, M. E., Salisbury-Glennon, J. D., Guarino, A., Reed, C. J., & Marshall, M. (2003). Situated self-regulation: modeling the interrelationships among instruction, assessment, learning strategies and academic performance. Educational Research and Evaluation, 9, 189-209. Rozendaal, J. S., Minnaert, A., & Boekaerts, M. (2003). Motivation and self-regulated learning in secondary vocational education: information-processing type and gender differences. Learning and Individual Differences, 13, 273-289. Säljö, R. (1979). 'Learning in the learner's perspective. I - Some common-sense conceptions’. University of Gothenburg. SBL. (2004). Competence requirements in pre-higher education [Electronic Version]. Retrieved January 27, 2010 from www.lerarenweb.nl. Schraw, G., & Dennison, R. S. (1994). Assessing metacognitive awareness. Contemporary Educational Psychology, 19, 460-475. Schunk, D. H. (2005). Commentary on self-regulation in school-contexts. Learning and Instruction, 15, 173-177. 182

References

Schunk, D. H., & Zimmerman, B. J. (1998). Self-regulated learning from teaching to selfreflective practice. New York: The Guildford Press. Severiens, S., Ten Dam, G., & Van Hout Wolters, B. (2001). Stability of processing and regulation strategies: Two longitudinal studies on student learning. Higher Education, 42, 437-453. Shulman, L. S. (1986). Those who understand: Knowledge growth in teaching. Educational Researcher, 15(2), 4-14. Singer, J. D., & Willett, J. B. (2003). Applied longitudinal data analysis. New York: Oxford University Press. Snijders, T. A. B. (1999). Multilevel analysis. An introduction to basic and advanced multilevel modeling. London: Sage publications. Strijbos, J., Meeus, W., & Libotton, A. (2007). Portfolio assignments in teacher education: A tool for self-regulating the learning process? International Journal for the Scholarship of Teaching and Learning, 1(2), 1-16. Ten Dam, G., & Blom, S. (2006). Learning through participation. The potential of school-based teacher education for developing a professional identity. Teaching and Teacher Education, 22, 647-660. Tillema, H. H. (2000). Belief change towards self-directed learning in student teachers: immersion in practice or reflection on action. Teaching and Teacher Education, 16, 575-591. Tillema, H. H. (2009). Assessment for learning to teach. Appraisal of practice teaching lessons by mentors, supervisors, and student teachers. Journal of Teacher Education, 60, 155-167. Tryggvason, M. T. (2009). Why is Finnish teacher education successful? Some goals Finnish teacher educators have for their teaching. European Journal of Teacher Education, 32, 369-382. Tynjälä, P. (1997). Developing education students’ conceptions of the learning process in different learning environments. Learning and Instruction, 7, 277-292. Tynjälä, P. (2008). Perspectives into learning at the workplace. Educational Research Review, 3, 130-154. Van Driel, J. H., Verloop, N., & De Vos, W. (1998). Developing science teachers’ pedagogical content knowledge. Journal of Research in Science Teaching, 35, 673695. Van Eekelen, I. M., Boshuizen, H. P. A., & Vermunt, J. D. (2005). Self-regulation in higher education teacher learning. Higher Education, 50, 447-471. Van Eekelen, I. M., Vermunt, J. D., & Boshuizen, H. P. A. (2006). Exploring teachers’ will to learn. Teaching and Teacher Education, 22, 408-423. Van Hout Wolters, B. (2000). Assessing active self-directed learning. In P. R. J. Simons, J. Van der Linden & T. Duffy (Eds.), New Learning (pp. 83-101). Dordrecht: Kluwer.

183

Van Hout Wolters, B. (2006, May). Leerstrategieën meten: Soorten meetmethoden en hun bruikbaarheid in onderwijs en onderzoek [Measuring learning strategies: Different measurement methods and their usability in education and research]. Paper presented at the Onderwijs Research Dagen, Amsterdam. Van Hout Wolters, B., Simons, P. R. J., & Volet, S. (2000). Active learning: Self-directed learning and independent work. In P. R. J. Simons, J. L. v. d. Linden & T. Duffy (Eds.), New Learning (pp. 21-37). Dordrecht: Kluwer Academic Publishers. Van Petegem, P., Donche, V., & Vanhoof, J. (2005). Relating pre-service teachers’ approaches to learning and preferences for constructivist learning environments. Learning Environments Research, 8, 309-332. Van Tartwijk, J., Driessen, E., Van der Vleuten, C., & Stokking, K. M. (2007). Factors influencing the successful introduction of portfolios. Quality in Higher Education, 13, 69-79. Veenman, M. V. J. (2005). The assessment of metacognitive skills: What can be learned from multi-method designs? In C. Artelt & B. Moschner (Eds.), Lernstrategien und metakognition: Implicationen für Forschung und Praxis (pp. 77-99). Münster: Waxmann. Vermetten, Y. J., Lodewijks, H. G., & Vermunt, J. D. (1999). Consistency and variability of learning strategies in different university courses. Higher Education, 37, 1-21. Vermetten, Y. J., Vermunt, J. D., & Lodewijks, H. G. (1999). A longitudinal perspective on learning strategies in higher education: Different viewpoints towards development. British Journal of Educational Psychology, 69, 221-242. Vermunt, J. D. (1996). Metacognitive, cognitive and affective aspects of learning styles and strategies: A phenomenographic analysis. Higher Education, 31, 25-50. Vermunt, J. D. (1998). The regulation of constructive learning processes. British Journal of Educational Psychology, 68, 149-171. Vermunt, J. D. (2005). Relations between student learning and personal and contextual factors and academic performance. Higher Education, 49, 205-234. Vermunt, J. D., & Minnaert, A. (2003). Dissonance in student learning patterns: when to revise theory? Studies in Higher Education, 28, 49-61. Vermunt, J. D., & Van Rijswijk, F. A. W. M. (1988). Analysis and development of students’ skill in self-regulated learning. Higher Education, 17, 647-682. Vermunt, J. D., & Verloop, N. (1999). Congruence and friction between learning and teaching. Learning and Instruction, 9, 257-280. Vermunt, J. D., & Verloop, N. (2000). Dissonance in students’ regulation of learning processes. European Journal of Psychology of Education, 15, 75-87. Vermunt, J. D., & Vermetten, Y. J. (2004). Patterns in student learning: Relationships between learning strategies, conceptions of learning and learning orientations. Educational Psychology Review, 16, 359-384. Volet, S., Summers, M., & Thurman, J. (2009). High-level co-regulation in collaborative learning. Learning and Instruction, 19, 128-143. 184

References

Weinstein, C. E., Schulte, A., & Palmer, D. (1987). LASSI: Learning and study strategies inventory. Clearwater: FL: H&H Publishing Westheimer, J. (2008). Learning among colleagues. Teacher community and the shared enterprise of education. In M. Cochran-Smith, S. Feiman-Nemser, D. J. McIntyre & K. E. Demers (Eds.), Handbook of research on teacher education (pp. 756-783). New York: Routledge. Willett, J. B. (1997). Measuring change: What individual growth modeling buys you. In E. Amsel & K. A. Renninger (Eds.), Change and development: Issues of theory, method, and application (pp. 213-243). Mahwah, NJ: Lawrence Erlbaum Associations. Winne, P. H. (1995a). Inherent details in self-regulated learning. Educational Psychologist, 30, 173-187. Winne, P. H. (1995b). Self-regulation is ubiquitous but it forms vary with knowledge. Educational Psychologist, 30, 223-228. Winne, P. H. (2005). Key issues in modeling and applying research on self-regulated learning. Applied Psychology: An International Review, 54, 232-238. Winne, P. H., & Jamieson-Noel, D. (2002). Exploring students’ calibration of self reports about study tactics and achievement. Contemporary Educational Psychology, 27, 551-572. Winne, P. H., & Perry, N. E. (2000). Measuring self-regulated learning. In M. Boekaerts, P. R. Pintrich & M. Zeidner (Eds.), Handbook of self-regulation (pp. 531-566). San Diego: Academic Press. Winters, F. I., Greene, J. A., & Costich, C. M. (2008). Self-regulation within computerbased learning environments: A critical analysis. Educational Psychology Review, 20, 429-444. Zanting, A., Verloop, N., & Vermunt, J. D. (2001). Student teachers’ beliefs about mentoring and learning to teach during teaching practice. British Journal of Educational Psychology, 71, 57-80. Zimmerman, B. J. (2000). Attaining self-regulation: A social cognitive perspective. In M. Boekaerts, P. R. Pintrich & M. Zeidner (Eds.), Handbook of self-regulation (pp. 13-39). San Diego: Academic Press. Zimmerman, B. J. (2002). Becoming a self-regulated learner: An overview. Theory Into Practice, 41, 64-70. Zimmerman, B. J. (2008). Investigating self-regulation and motivation: Historical background, methodological developments, and future prospects. American Educational Research Journal, 45(1), 166-183. Zimmerman, B. J., & Schunk, D. H. (2001). Self-regulated learning and academic achievement. Theoretical perspectives (2nd ed.). Mahwah: Lawrence Erlbaum Associates.

185

Zwart, R., Wubbels, T., Bolhuis, S., & Bergen, T. C. M. (2008). Teacher learning through reciprocal peer coaching: An analysis of activity sequences. Teaching and Teacher Education, 24, 982-1002.

186

specifieke

bepaalde

opleiding

kunnen

volgen

dio

regulatietypes

mate alleen

baan

beschrijven

veel

gebruikt

passief eigen

stellen

alle

goed

reflectie

context

studie

reguleren

docenten

gericht duidelijke

worden

componenten zelf

regulatieactiviteiten

duale universiteit

regulatiepatroon

prospectief

eerder

gestructureerd

betekent

minder

meerdere begeleiden

logboek

krijgen

stagevariant

ontwikkeling baanvariant

portfolio volgende

richting

reproductiegericht

leerproces

verschillen

daarnaast

ontwikkelen

sturen

gereguleerde

inzicht

ervaringen

belangrijk

lesgeven

hoe basis

eenjarige

gedurende

leeromgeving

twee betekenisgericht gemeten

proefschrift

vraag

zelfregulerend

versie

longitudinale

gevonden

manier andere

evaluatie

theorie

kwadranten

behulp

dynamische

retrospectief

beoordeling

nodig resultaten

onderzoek

praktijk

hoofdstuk

stabiele

focus

monitoring

variatie

lerarenopleiding

staat

scores

leren

structuur

maken

beide verwachting

onderscheiden

eind

leerrapport

studenten

combinatie

leerconcepties aangetoond kwaliteit verschillende verder

regulatie aard

relatie

actief

individuele

belang

algemene

varianten

keuzes

zien

zes

tussen plaatsvonden

onafhankelijk

dimensies

leerervaringen laten

validiteit

Samenvatting (Dutch summary) Algemene inleiding Het doel van het onderzoek dat in dit proefschrift is beschreven, is om meer inzicht te verkrijgen in de aard en de ontwikkeling van zelfregulerend leren van docenten in een duale opleiding. In een duale opleiding leren docenten in opleiding (dio’s) parallel in twee contexten: op de universiteit en in de praktijk. Deze eenjarige universitaire lerarenopleiding is slechts een kortdurende interventie in de gehele professionele ontwikkeling van docenten. Het is daarom één van de taken van deze lerarenopleiding om dio’s te begeleiden bij de ontwikkeling van de concepties en vaardigheden die nodig zijn om zich ook na de opleiding actief te blijven ontwikkelen. Om dio’s voor te bereiden op actief levenslang leren wordt al gedurende de opleiding van hen verwacht dat zij hun eigen leerproces gaan aansturen. Zo moeten ze een persoonlijk ontwikkelingsplan maken, leerbehoeftes formuleren, leerervaringen uit meerdere contexten combineren en zichzelf beoordelen. Aangezien dio’s in grote mate zelf verantwoordelijk zijn voor hun eigen leerproces, wordt de kwaliteit van leren van dio’s voor een groot gedeelte bepaald door de capaciteit om hun eigen leren te sturen. Momenteel is er nog weinig bekend over hoe en in welke mate dio’s hun eigen leren reguleren, evenals hoe zij zich hierin ontwikkelen gedurende de opleiding. De assumptie is dat een beter begrip van de aard en ontwikkeling van zelfregulerend leren van dio’s nodig is om een krachtige lerarenopleiding te ontwerpen die hun ontwikkeling als lerende ondersteunt en stimuleert. De hoofdvraag van dit proefschrift is dan ook: Wat is de aard en ontwikkeling van zelfregulerend leren van dio’s in een duale opleiding?

Onderzoek naar leren lesgeven Onderzoek naar wat docenten leren, laat zien dat zij behalve leren om te denken, weten, voelen en handelen als een docent, ook moeten leren om te leren als een docent. Een kenmerk van een expert-docent is dat hij niet alleen in staat is om bepaalde routines op te bouwen, maar ook om deze te blijven ontwikkelen. De vraag hoe docenten leren lesgeven heeft in het afgelopen decennium dan ook steeds meer aandacht gekregen. Het leren van docenten 189

wordt vaak beschreven als een complex, multidimensioneel, idiosyncratisch en contextspecifiek proces. Onderzoek heeft aangetoond dat docenten meer dan vijf jaar nodig hebben om zich te ontwikkelen tot een expert-docent. In de dynamische omgeving van de dagelijkse lespraktijk is het voor docenten vaak lastig om hun ontwikkeling tot expert-docent goed te blijven sturen. Lerarenopleidingen zijn in het verleden vaak bekritiseerd omdat zij de studenten niet goed bleken voor te bereiden op de dagelijkse praktijk van het lesgeven. De beginnende docenten ervoeren een grote praktijkschok wanneer zij na hun theoretische opleiding gingen lesgeven. Deze kritiek heeft geleid tot programma’s waarin theorie en praktijk meer geïntegreerd worden aangeboden, zoals in een duale opleiding het geval is. Uit onderzoek blijkt echter, dat dio’s verschillen in de mate waarin zij in staat zijn om succesvol van deze praktijkervaringen te leren en hun leren aan te sturen. Dit laat zien dat zelfregulerend leren niet alleen een gewenste uitkomst is van de initiële opleiding, maar ook een essentiële vaardigheid om gedurende de opleiding te leren van de combinatie van theorie en praktijk.

Onderzoek naar zelfregulerend leren De huidige brede definitie van zelfregulerend leren omvat zowel een stabiele component die de kennis beschrijft die iemand over zijn leren heeft (metacognitieve kennis), als een meer dynamische component die de regulatie van het leren omvat (metacognitieve regulatie). Ook al zijn deze twee componenten onderdeel van hetzelfde concept, ze zijn en worden veelal apart bestudeerd vanuit verschillende perspectieven. In theorieën die de dynamische regulatiecomponent beschrijven, wordt ervan uitgegaan dat studenten op een actieve, constructieve manier aan hun leren deelnemen en dat alle studenten in principe in staat zijn hun eigen leren te sturen. Het leerproces wordt aangestuurd door het vastgestelde leerdoel (Pintrich, 2000). De regulatie van het leerproces bestaat uit verschillende opeenvolgende fases: (1) planning en doelen stellen, (2) monitoring en (3) controle, en (4) reflectie en evaluatie. Er is veel onderzoek gedaan naar hoe docenten zelfregulatie van leren bij hun leerlingen kunnen stimuleren, maar weinig naar hoe zij hun eigen leren aansturen. Oorspronkelijk was onderzoek naar metacognitieve kennis voornamelijk gefocust op de feitelijke kennis die studenten hadden over hun leren. Studies naar leeroriëntaties en leerstijlen lieten het belang zien van het geïntegreerd bestuderen van leerconcepties, motivatie en voorkeuren voor 190

Samenvatting

leer- en regulatiestrategieën. Wanneer deze leeroriëntaties en leerstijlen op een niet-situatiespecifieke manier worden gemeten, kunnen deze concepten goed gebruikt worden om iemands algemene benadering van het leren te beschrijven. Onderzoeken hebben herhalend aangetoond dat de dimensies reproductiegericht en betekenisgericht leren onderscheidend zijn voor verschillende benaderingen van studenten van hun leerproces. Onderzoek dat gedaan is naar leerconcepties en leeroriëntaties van dio’s heeft zich veelal gericht op hun manieren van leren in een schoolse context. De studie van Oosterheert en Vermunt (2001) toonde aan dat ook bij dio’s in een duale opleiding verschillende leeroriëntaties te onderscheiden zijn. Op basis van hun onderzoek hadden zij de verwachting dat dio’s zich met de juiste begeleiding kunnen ontwikkelen van overlevingsgericht of reproductiegericht naar betekenisgerichte manier van leren.

Focus en context van dit onderzoek Om een volledig beeld te krijgen van zelfregulerend leren van dio’s, worden in dit onderzoek beide componenten van zelfregulerend leren onderzocht. In dit proefschrift verwijzen we dan ook met de term zelfregulerend leren naar de combinatie van de kenniscomponent en de regulatiecomponent. Het concept leeroriëntatie is gebruikt om de meer stabiele kenniscomponent te onderzoeken. De dynamische regulatiecomponent is in dit proefschrift geoperationaliseerd als de combinatie van regulatieactiviteiten die dio’s gebruiken in concrete leerervaringen. We gebruiken de termen regulatie, zelfregulatie, zelfregulatie van leren als we naar deze dynamische component refereren. De hoofdvraag naar de aard en ontwikkeling van zelfregulerend leren van dio’s is opgesplitst in vier deelvragen die in de verschillende hoofdstukken worden beantwoord: 1. Wat is de aard van de regulatie van leren van dio’s in een duale opleiding? (Hoofdstuk 2) 2. Wat zijn verschillen in regulatiepatronen en leerconcepties van dio’s in een duale opleiding en hoe zijn deze aan elkaar gerelateerd? (Hoofdstuk 3) 3. In welke mate is een gestructureerd logboek dat meerdere leerervaringen beschrijft, een valide en betrouwbaar instrument om verschillende kwaliteiten van regulatie van leren van dio’s in een duale opleiding te meten en te onderscheiden? (Hoofdstuk 4)

191

4. Hoe veranderen regulatie van leren en leeroriëntaties van dio’s gedurende een eenjarige duale opleiding en wat zijn hierin verschillen tussen twee opleidingsvarianten? (Hoofdstuk 5) Deze vragen zijn onderzocht in een eenjarige Nederlandse universitaire lerarenopleiding. Deze opleiding bestaat uit twee varianten: een baan- en een stagevariant. De studenten met een baan starten vanaf het begin van de opleiding als docent op een middelbare school. Studenten die de stagevariant volgen, beginnen met een triostage waarin de praktijkcomponent meer geleidelijk is opgebouwd: ze observeren ervaren docenten en mede-dio’s en geven hun eerste (deel)lessen. In de tweede helft van de opleiding volgen zij een liostage waarin ze meer verantwoordelijkheden als docent hebben. Het instituutsgedeelte van de opleiding is voor beide varianten gelijk. Alle dio’s volgen wekelijks algemeen didactische en vakdidactische bijeenkomsten. Dio’s maken een persoonlijk ontwikkelingsplan waarin ze in overleg met hun opleider hun programma samenstellen. Zij houden hun ontwikkeling bij aan de hand van verschillende competenties in een portfolio. Zelfregulerend leren is een belangrijk onderdeel van de SBL1-competentie “reflectie en ontwikkeling”. In het portfolio nemen dio’s ook bepaalde producten op, zoals lesplannen, observaties, video’s en opdrachten. De beoordeling aan het eind van de opleiding is gebaseerd op het portfolio en een eindgesprek met hun lerarenopleider.

De aard van regulatie van leren van dio’s in een duale opleiding In Hoofdstuk 2 staat de aard van regulatie van dio’s in beide contexten van een duale leeromgeving centraal. Voorgaand onderzoek naar regulatie van leren is vooral gericht geweest op het leren van studenten in schoolse of experimentele omgevingen. Dit onderzoek heeft als doel om de variatie in regulatieactiviteiten van dio’s en onderliggende structuur in een duale leeromgeving in kaart te brengen. Hiertoe is aan 28 dio’s gevraagd om zes leerervaringen te beschrijven in een logboek met open vragen. Uitgangspunt bij de constructie van dit logboek, Leerrapport genaamd, was het raamwerk van Pintrich (2000), die verschillende fasen onderscheidt in het regulatieproces. De vragen in het logboek refereerden aan deze fasen (planning en doelen 1

Stichting Beroepskwaliteit Leraren

192

Samenvatting

stellen, monitoring en controle, en reflectie en evaluatie). Inhoudsanalyse van de 133 verzamelde Leerrapporten liet zien dat deze variëteit beschreven kon worden aan de hand van acht variabelen met vijf tot zeven categorieën per variabele. De variabelen waren: doel-oriëntatie, succesverwachting, leerstrategie, leerstrategiebewaking, monitoring van de leerresultaten, reflectie op de leeruitkomst, zelfevaluatie van het leerproces, en voorbereiding op een volgende leerervaring. Om de onderliggende structuur en de relatie tussen al deze categorieën te achterhalen is multiple correspondence analysis toegepast. Deze analyse toonde een tweedimensionale structuur in de data aan. Deze twee dimensies onderscheidden (1) passieve van actieve regulatie en (2) prospectieve van retrospectieve regulatie. Een passief gereguleerde leerervaring wordt gekarakteriseerd door gebrek aan argumentatie voor de keuzes die de dio maakt gedurende de leerervaring. Daarnaast is vaak iemand anders aanwezig die onderdelen van het leerproces stuurde. Bij een actief gereguleerde leerervaring maakt de dio meer weloverwogen keuzes voor een leerdoel, strategie en reflecteert hij of zij op een diepere manier op wat er is geleerd, het leerproces en zijn of haar rol hierin. Bij een prospectief gereguleerde ervaring ligt de focus van de regulatie op de beginfase van de leerervaring. Dit betekent dat de dio actief doelen stelt en een strategie kiest, maar het monitoren, evalueren en reflecteren krijgt minder aandacht. Een retrospectief gereguleerde leerervaring is vaak een ongeplande leerervaring waarbij dus geen actieve regulatie vooraf heeft plaatsgevonden. De focus van de regulatie ligt meer op de laatste fase van de leerervaring: de dio laat een diepgaandere manier van monitoring, evaluatie en reflectie zien. Deze tweede dimensie impliceert dat leerervaringen die starten als reactief, ongepland en geen voorafgestelde leerdoelen hebben, toch actief gereguleerd kunnen worden, zij het op een meer retrospectieve manier. Deze kleinschalige studie was er ook op gericht om na te gaan of er verschil is tussen de aard van de regulatie in leerervaringen die op de universiteit of in de praktijk, plaatsvonden. Deze resultaten laten zien dat leerervaringen die plaatsvonden in de praktijk significant actiever werden gereguleerd dan leerervaringen die op de universiteit plaatsvonden. Er zijn geen verschillen in prospectiviteit en retrospectiviteit van de leerervaringen in beide contexten. Hoewel het leren op de werkplek vaak wordt gezien als minder weloverwogen en meer reactief, laten deze resultaten zien dat de context van de praktijk toch meer actieve regulatie uitlokte dan de context van de universiteit. 193

Individuele verschillen in zelfregulerend leren van dio’s in een duale opleiding In Hoofdstuk 3 van het proefschrift verschuift de focus van variatie op het niveau van leerervaringen naar variatie in zelfregulerend leren op het niveau van de individuele dio. Om variatie in regulatie van dio’s te beschrijven is dezelfde verzameling Leerrapporten gebruikt als beschreven in Hoofdstuk 2. De meer stabiele component van zelfregulerend leren is gemeten met behulp van interviews. In dit hoofdstuk is het concept leerconcepties gebruikt om deze component te beschrijven, aangezien in de interviews alleen de algemene conceptie van de dio’s van het leren lesgeven is bevraagd en niet andere aspecten van hun leeroriëntatie, zoals de specifieke leer- en regulatiestrategieën. Van de 28 participerende dio’s bevonden 14 dio’s zich aan het begin van de opleiding en 14 dio’s aan het eind ervan. Analyse van de Leerrapporten op individueel niveau heeft laten zien dat dio’s zich niet beperken tot één type regulatie, maar in de zes leerervaringen meerdere regulatietypes lieten zien. Vier regulatietypen konden worden onderscheiden: actief prospectief, actief retrospectief, passief prospectief en passief retrospectief. Om hun regulatiepatroon te beschrijven zijn de zes leerervaringen gevisualiseerd in de vier kwadranten van de onderliggende structuur. Van de 21 dio’s van wie een regulatiepatroon kon worden vastgesteld, hadden vijf dio’s een veelzijdige manier van reguleren. Dat wil zeggen, dat zij leerervaringen in alle vier de kwadranten lieten zien. De andere dio’s hadden een duidelijkere focus in hun regulatie en zijn geclassificeerd als dio’s met een actief, prospectief of retrospectief regulatiepatroon. Voor de categorisatie van de leerconcepties van de dio’s is het onderscheid tussen reproductiegericht leren en betekenisgericht leren gebruikt, zoals beschreven door Oosterheert en Vermunt (2001). Acht van de 28 dio’s lieten een betekenisgerichte leerconceptie zien. Opvallend was dat van deze acht dio’s, slechts één een ervaren dio was. De begrippen consonantie en dissonantie zijn gebruikt om de relatie tussen de leerconceptie en het regulatiepatroon te beschrijven. Een betekenisgerichte leerconceptie gecombineerd met een actief regulatiepatroon, en een reproductiegerichte leerconceptie met een passief regulatiepatroon worden als theoretisch consonant gezien. Op basis hiervan hadden drie dio’s een consonant regulatiepatroon met hun leerconceptie en zes dio’s een dissonant patroon. De overige dertien dio’s konden niet eenduidig als consonant of dissonant worden 194

Samenvatting

geclassificeerd, omdat zij in hun regulatiepatroon zowel actieve als passieve regulatie lieten zien. De resultaten laten zien dat in een duale leeromgeving meer combinaties voorkomen van regulatiepatronen en leerconcepties dan is gesuggereerd in eerder onderzoek. Om meer inzicht te krijgen in de relatie tussen deze twee componenten van zelfregulerend leren is onderzoek met een grotere steekproef noodzakelijk. Daarnaast lijkt het van belang om naast leerconcepties ook voorkeuren voor leer- en regulatieactiviteiten erbij te betrekken.

De ontwikkeling en evaluatie van een instrument om regulatie van leren te meten De grote variëteit in regulatietypes binnen dio’s zoals in Hoofdstuk 3 beschreven laat zien dat het bestuderen van één leerervaring niet voldoende is om de kwaliteit van regulatie van een dio in een duale opleiding te beschrijven. De ongestructureerde Leerrapporten zoals gebruikt in de eerste studie zijn echter te arbeidsintensief voor zowel de dio’s als de onderzoekers om grootschalig onderzoek naar zelfregulerend leren van dio’s mogelijk te maken. Hoofdstuk 4 beschrijft de ontwikkeling en evaluatie van het Gestructureerd Leerrapport om op een valide en betrouwbare manier de kwaliteit van regulatie te beschrijven op basis van meerdere leerervaringen. Dit instrument bestaat uit één open vraag en zeven meerkeuzevragen. De opties van de meerkeuzevragen waren gebaseerd op de antwoorden gegeven in de open versie van het Leerrapport. De gestructureerde versie is zes keer afgenomen bij 81 dio’s op drie momenten gedurende de opleiding. Een vergelijking tussen de open versie en de gestructureerde versie liet zien dat beide versies een zelfde hoge constructvaliditeit en betrouwbaarheid hebben. Dit betekent dat het vervangen van open vragen door gesloten vragen geen negatieve invloed heeft gehad op de kwaliteit. Wel is hierdoor het gebruiksgemak van het instrument voor de lerarenopleiding verhoogd. Het gebruik van het instrument door de dio’s is geanalyseerd met behulp van een kleine evaluatievragenlijst. In deze vragenlijst gaf de meerderheid van de dio’s aan dat ze veel geleerd hadden van deelname aan het onderzoek. Slechts een enkele dio had de tijdsinvestering als groot ervaren. Om een indicatie te krijgen van de predictieve validiteit en concurrente validiteit zijn extra metingen verricht. Voor de predicitieve validiteit zijn de scores op de regulatiedimensies van de dio’s vergeleken met de beoordeling 195

van hun verschillende competenties door hun opleider. Hier bleek geen verband tussen te bestaan. Dat betekent dat dio’s die op een bepaalde manier reguleren niet met een hogere of lagere beoordeling de opleiding verlaten. De scores op de regulatiedimensies zijn ook gerelateerd aan de scores op de Inventory Learning to Teach Process (ILTP, Oosterheert, Vermunt & Denessen, 2002), om een indicatie te krijgen voor de concurrente validiteit. De ILTP meet de leeroriëntaties van dio’s, waarvan regulatie ook een onderdeel is. Ook hier werd geen relatie gevonden: dio’s met een bepaalde leeroriëntatie bleken niet te verschillen in hun scores op de regulatiedimensies. Verder onderzoek is nodig om de predictieve en concurrente validiteit vast te stellen.

De ontwikkeling van zelfregulerend leren van dio’s in een duale opleiding Zoals beschreven in de voorgaande hoofdstukken blijken dio’s te verschillen in de mate waarin zij actief hun leerproces reguleren en waarin zij dit meer prospectief of meer retrospectief doen. Daarnaast heeft onderzoek laten zien dat dio’s verschillende leeroriëntaties hebben die variëren in de mate waarin ze onafhankelijk en betekenisgericht leren. Gezien het groeiende lerarentekort heeft een steeds groter deel van de studenten al tijdens de opleiding een baan en daarmee alle verantwoordelijkheden van een reguliere docent. Dit in tegenstelling tot de stagevariant, waarin een geleidelijke opbouw van het praktijkgedeelte van de opleiding wordt gewaarborgd. Onderzoekers, maar ook lerarenopleiders hebben hun zorgen geuit over varianten van lerarenopleidingen waarin de praktijkcomponent een grote plaats inneemt. Deze dio’s zouden te weinig tijd hebben voor reflectie op hun ervaringen en om leerervaringen goed vorm te geven. Daarom staat in Hoofdstuk 5 de volgende vraag centraal: Hoe veranderen regulatie van leren en leeroriëntaties van dio’s gedurende een eenjarige duale opleiding en wat zijn hierin verschillen tussen twee opleidingsvarianten? Op basis van de literatuur was de verwachting dat dio’s gedurende de opleiding meer actief zouden gaan reguleren en zich zouden ontwikkelen in de wenselijke richting van een onafhankelijke betekenisgerichte manier van leren. Daarnaast was de verwachting dat dio’s die een opleidingsvariant met een geleidelijk opbouw volgden, hun leren op een actievere en meer betekenisgerichte manier zouden vormgeven.

196

Samenvatting

Deze onderzoeksvragen zijn onderzocht in een longitudinale studie door op drie momenten gedurende de opleiding de leeroriëntatie en regulatie van leren in verschillende leerervaringen van dio’s te meten. Op het eerste moment van afname waren de dio’s drie maanden met de opleiding bezig. De volgende twee meetmomenten volgden respectievelijk drie en zes maanden later. Eenentachtig dio’s (90%) hebben meegewerkt aan het onderzoek, waarvan 47 de stagevariant deden en 34 dio’s de baanvariant. Dio’s met een baan gaven ongeveer twee keer zoveel lessen, besteedden meer tijd aan het praktijkgedeelte van de opleiding en hadden ook meer leservaring voordat ze aan de opleiding begonnen. De leeroriëntatie is op elk meetmoment gemeten met de ILTP. Met behulp van de scores op deze vragenlijst kunnen dio’s worden ingedeeld in één van de vier onderscheiden leeroriëntaties: inactief/overlevingsgericht, reproductiegericht, afhankelijk betekenisgericht en onafhankelijk betekenisgericht. Met het Gestructureerde Leerrapport hebben we op elk afnamemoment de regulatieactiviteiten van zelfgerapporteerde leerervaringen gemeten. Per meetmoment is aan de dio’s gevraagd om zes verschillende leerervaringen uit beide contexten (universiteit en school) te rapporteren. De data zijn verwerkt met behulp van multiple correspondence analysis, longitudinale multilevelanalyse en patroonanalyses. De resultaten laten zien dat dio’s door het jaar heen passiever werden in hun manier van reguleren. Daarnaast reguleerden dio’s in het begin het meest prospectief, wat inhoudt dat ze actief bezig waren om leerervaringen aan te sturen, maar dat ze naderhand deze ervaringen op een oppervlakkige manier verwerkten. Op het tweede meetmoment, reguleerden de dio’s het meest retrospectief: De leerervaringen waren vaker ongepland: de actieve regulatie focuste op het actief monitoren, evalueren en reflecteren op de leerervaring. Aan het eind van de opleiding werden de dio’s weer prospectiever in hun regulatie. Er was geen verschil tussen de twee varianten in de mate van passiviteit en retrospectiviteit van de regulatie. In termen van leeroriëntaties behield ongeveer een derde dezelfde leeroriëntatie en een derde veranderde richting onafhankelijk betekenisgericht leren. Een vijfde van de dio’s veranderde ook van leeroriëntatie, maar dan in tegengestelde richting. Dio’s uit de baanvariant scoorden significant hoger op de ILTP-schalen “actief relateren theorie aan praktijkervaring” en “ideeën/visie ontwikkelen door discussie” en lager op de leerconceptie “lesgeven en uitproberen” en “zorgen hebben over negatieve leservaringen”. Deze dio’s hadden ook vaker een betekenisgerichte leeroriëntatie. Dit verschil was groter aan het eind van de 197

opleiding. In termen van ontwikkeling veranderden dio’s uit de baanvariant minder vaak van leeroriëntatie. Dio’s uit de stagevariant wisselden vaker van oriëntatie, zowel richting onafhankelijk betekenisgericht leren, als daar vandaan. Deze bevindingen zijn niet in overeenstemming met onze verwachtingen. Het feit dat studenten passiever worden, betekent niet noodzakelijkerwijs dat ze ook minder in staat zijn actief te reguleren. Het kan ook duiden op een consolidatiefase in hun professionele ontwikkeling. Verklaringen voor gebrek aan ontwikkeling richting onafhankelijk betekenisgericht leren kan duiden op een bepaalde vorm van weerstand bij dio’s om hun huidige referentiekader los te laten. Verder tonen de resultaten aan dat het verwachte negatieve effect van de tijdsintensieve baanvariant op de ontwikkeling van leren lesgeven niet kon worden bevestigd. In tegendeel, over het algemeen laten dio’s met een baan een meer betekenisgerichte manier van leren zien. Verklaringen voor deze resultaten zijn speculatief. Het zou kunnen zijn dat volledige verantwoordelijkheid in een baan ook dio’s meer stimuleert om zelf actief na te denken over hun manier van lesgeven en om zich daarin actief te ontwikkelen, waarbij stagiaires zich meer afhankelijk van de keuzes van hun mentor opstellen. Verder onderzoek is nodig om de gevonden verschillen te verklaren. Deze studie laat zien dat het geven van veel verantwoordelijkheid voor het eigen leren wellicht een noodzakelijke, maar zeker geen voldoende voorwaarde is om het zelfregulerend vermogen bij dio’s te ontwikkelen. Dit betekent dat er meer gerichte interventies nodig zijn om docenten te helpen om de concepties en vaardigheden te ontwikkelen die nodig zijn voor levenslang leren als docent.

Algemene conclusie en discussie In Hoofdstuk 6 zijn de bevindingen samengevat en bediscussieerd. De studies in dit proefschrift hebben antwoorden gevonden op de vragen naar de aard van de regulatie van dio’s, individuele verschillen in kwaliteit van regulatie en de ontwikkeling van dio’s in zelfregulerend leren gedurende een duale opleiding. Verder heeft dit onderzoek geleid tot de ontwikkeling van het Gestructureerd Leerrapport, waarmee we op een valide en betrouwbare manier verschillende kwaliteiten van regulatie in een duale opleiding kunnen beschrijven en onderscheiden. Uit de studies kunnen de volgende conclusies 198

Samenvatting

worden getrokken: De aard van de regulatie van leerervaringen van dio’s kan worden beschreven aan de hand van de dimensies actief versus passief, en prospectief versus retrospectief. Dit levert een indeling in vier kwadranten op die de verschillende regulatietypes van leren van dio’s in een duale opleiding beschrijven. Daarnaast toonde dit onderzoek grote variëteit aan binnen dio’s in type regulatie dat zij toepassen in hun verschillende leerervaringen. Leerervaringen die plaatsvonden in de context van de schoolpraktijk, werden door dio’s gemiddeld actiever gereguleerd dan leerervaringen die plaatsvonden op de universiteit. Hun individuele regulatiepatronen bestaan dan ook uit meerdere regulatietypen. Deze regulatiepatronen van dio’s waren in veel gevallen niet (geheel) consonant met hun leerconceptie. De longitudinale studie heeft laten zien dat dio’s gedurende de opleiding hun leren meer passief gingen reguleren. Halverwege de opleiding waren de dio’s het meest retrospectief in hun regulatie. Een derde van de dio’s behield gedurende de lerarenopleiding dezelfde leeroriëntatie en een derde ontwikkelde zich richting onafhankelijk betekenisgericht leren. De resultaten toonden ook aan dat dio’s uit de baanvariant vaker een betekenisgerichte leeroriëntatie hadden. Dit verschil was groter aan het eind van de opleiding. Dio’s uit de baanvariant veranderden wel minder vaak van leeroriëntatie. Deze resultaten dragen bij aan ons begrip over de aard en ontwikkeling van zelfregulerend leren van dio’s. Wat betreft de aard van zelfregulerend leren hebben de in dit proefschrift beschreven studies laten zien dat binnen actieve regulatie van dio’s zowel actieve prospectieve als retrospectieve regulatie kan worden onderscheiden. Actieve prospectieve regulatie door bijvoorbeeld het stellen van leerdoelen die het leerproces sturen, is zeker belangrijk, maar niet de enige manier om actief te reguleren. Daarnaast heeft de studie laten zien dat dio’s om hun leren te reguleren vaak ook werkdoelen” of gedragsdoelen stellen en niet alleen leerdoelen. Bovendien hebben deze studies aangetoond dat het veelgebruikte onderscheid tussen zelf- en externe regulatie niet zo relevant is voor het leren in een duale leeromgeving. Belangrijker is de manier waarop de dio actief zijn of haar leerproces reguleert, hiervoor anderen weet te benutten en zo van zijn of haar omgeving een leeromgeving weet te maken. Over de ontwikkeling van zelfregulerend leren heeft dit onderzoek laten zien dat zowel de stabiele als de dynamische component van zelfregulerend leren veranderbaar zijn. De studies hebben ook laten zien dat deze verandering niet automatisch in de gewenste richting gaat. Dit hoeft niet 199

per se te betekenen dat het dio’s ook aan capaciteit ontbreekt om hun eigen leren aan te sturen. Het kan ook zijn dat dio’s niet de noodzaak voelen om dit te doen of hier niet de gelegenheid toe hebben. Dit proefschrift heeft ook een aantal vragen onbeantwoord gelaten. Zo hebben we in de beschreven studies niet een eenduidige manier kunnen vinden waarop de twee componenten van zelfregulerend leren aan elkaar gerelateerd zijn. Nu is ook in eerder onderzoek regelmatig geen relatie gevonden tussen zelfregulerend leren gemeten met een algemene vragenlijst en gemeten met situatiespecifiek instrument. We maken hieruit op dat de twee componenten van zelfregulerend leren eerder als complementair moeten worden gezien dan als overlappend. Verder onderzoek moet uitwijzen onder wat voor omstandigheden een dio specifieke combinaties van leeroriëntaties en regulatiepatronen vertoont. Een andere onbeantwoorde vraag is op welke manier zelfregulerend leren bij dio’s effect heeft op hun prestaties. Zo hebben we geen relatie gevonden tussen de kwaliteit van regulatie en het competentieniveau van de dio aan het einde van de opleiding zoals beoordeeld door de opleider met een rubric-systeem. Het zal nodig zijn om dio’s langer te volgen om na te gaan op welke manier zelfregulerend leren voor dio’s het verschil maakt in hun professionele ontwikkeling als docent. De studies hebben wel aangetoond dat er duidelijke verschillen zijn tussen dio’s in hun manier van leren gedurende de opleiding. De vraag is nog wat deze verschillen precies veroorzaakt. In ons onderzoek hebben we laten zien dat de context waarin geleerd wordt, het moment in de opleiding en de variant van de opleiding die gevolgd wordt, van invloed zijn op hoe dio’s leren. Daarmee blijft echter nog een groot gedeelte van de variantie tussen dio’s onverklaard. Bovendien wezen de gevonden verschillen vaak in een andere richting dan op basis van de literatuur kon worden verwacht. Vervolgonderzoek moet aantonen op welke manier de verschillende persoons- en contextfactoren het zelfregulerend leren van dio’s beïnvloeden. Het onderzoek kent ook andere beperkingen. Zo hebben we ervoor gekozen om dio’s zelf leerervaringen te laten selecteren om over te rapporteren. We hebben geen informatie verzameld over hoe zij deze selectie hebben gemaakt en dit kan de resultaten hebben beïnvloed. Daarnaast hebben we het onderzoek op slechts één lerarenopleiding uitgevoerd en hebben we niet meer dan drie meetmomenten opgenomen in de longitudinale studie. Een grotere hoeveelheid data had wellicht meer inzicht in de verschillen in 200

Samenvatting

ontwikkeling van dio’s kunnen geven. Aan de andere kant had een extra kleinschaligere studie of andere analysetechnieken meer vragen kunnen beantwoorden met betrekking tot de specifieke kenmerken van situaties waarin dio’s op een bepaalde manier leren. Ondanks deze beperkingen heeft dit onderzoek aan aantal duidelijke implicaties voor de praktijk. Ten eerste heeft dit onderzoek een duidelijkere invulling gegeven aan de SBL-competentie “reflectie en ontwikkeling”. Het onderzoek maakt ook duidelijk dat het ontwerp van de lerarenopleiding niet alleen dio’s de mogelijkheid tot zelfregulerend leren moet geven, maar ook de mogelijkheid om hun zelfregulerend leren te ontwikkelen. In de longitudinale studie is duidelijk geworden dat de ontwikkeling in hun manier van leren verschillend is voor dio’s in de baan- en stagevariant. Dio’s in de stagevariant wisselden vaker van leeroriëntatie, maar ook vaker in ongewenste richting. Daarom is het van belang om de wijzigingen in leervoorkeuren goed te monitoren bij studenten in de stagevariant, bijvoorbeeld wanneer zij wisselen van school. Momenteel worden veelal reflectiemodellen gehanteerd in universitaire lerarenopleidingen om dio’s te begeleiden bij hun leren. Ook al is reflectie een sleutelelement voor het leren van ervaringen, een exclusieve focus op het leren door terug te kijken op ervaringen is te beperkt om op een actieve en betekenisgerichte manier te leren. Het is van groot belang dat dio’s ook leren om zelf leerervaringen te creëren op basis van hun ideeën en theorie. Het portfolio wordt veelal gezien als een instrument waarmee dio’s hun leerproces leren aan te sturen. Eerder onderzoek heeft al aangetoond dat het bijhouden van een portfolio niet automatisch leidt tot de ontwikkeling van het vermogen om zelfregulerend te leren. Dit betekent dat er een meer gerichte didactiek ontwikkeld moet worden om het zelfregulerend leren van dio’s te helpen ontwikkelen. Een startpunt voor het begeleiden van dio’s bij hun ontwikkeling tot zelfregulerend lerende docenten is om ze bewust te maken van hoe ze leren. Zowel de ILTP, als het Gestructureerd Leerrapport kunnen hiertoe gebruikt worden. De uitkomsten van deze instrumenten kunnen als startpunt dienen voor de discussie over hoe je als docent leert en wat goede manieren van leren zijn. Daarnaast is het van belang dat dio’s inzicht krijgen in alternatieve manieren van leren. Zij kunnen alleen een weloverwogen keuze maken voor een manier van leren, wanneer ze weten wat de mogelijkheden zijn. Hiervoor kan het behulpzaam zijn om leerervaringen te bespreken met een mede-dio die 201

een andere manier van leren heeft. Daarnaast zullen studentspecifieke strategieën nodig zijn om dio’s te begeleiden bij hun ontwikkeling, afgestemd op de context en persoonlijke achtergrond van de dio. De lerarenopleider speelt hierin een cruciale rol door de specifieke behoeften van de dio te identificeren en met hen de juiste aanpak te kiezen. Tot slot, dit onderzoek biedt verschillende mogelijkheden tot vervolgonderzoek naar het leren van dio’s in een duale opleiding. Om te begrijpen waarom dio’s op een bepaalde manier leren is het van belang om meer diepgaand onderzoek te doen naar de specifieke factoren die invloed hebben op het leerproces van de dio. Hierbij kan naast persoonsgebonden factoren ook worden gedacht aan factoren op de praktijkschool, die de manier van leren van de dio kunnen beïnvloeden. Zoals hierboven beschreven, is een belangrijke vervolgstap om verschillende didactieken te ontwikkelen en te onderzoeken, die zelfregulerend leren van dio’s stimuleren. Het vertalen van onderzoeksuitkomsten en ideeën naar de praktijk is iets waar elke docent en dus ook lerarenopleiders goede ondersteuning bij kunnen gebruiken en waar dan ook meer onderzoek op gericht zou mogen zijn.

202

Publications Manuscripts submitted for publication Endedijk, M., Vermunt, J.D., Verloop, N., & Brekelmans, M. (under revision). The nature of student teachers’ regulation of learning in a dual learning environment. Endedijk, M., Brekelmans, M., Verloop, N., & Vermunt, J.D. (under revision). Student teachers’ self-regulated learning: An examination of regulation patterns and conceptions of learning in a dual learning programme. Endedijk, M., Brekelmans, M., Den Brok, P., & Vermunt, J.D. (submitted). Development and evaluation of an instrument to measure student teachers’ regulation of learning in dual learning programmes. Endedijk, M., Vermunt, J.D., Meijer, P.C., & Brekelmans, M. (submitted). Student teachers’ development in self-regulated learning throughout a dual learning programme.

Papers Endedijk, M., Meijer, P., Vermunt, J., & Brekelmans, M. (2009). De ontwikkeling van leeroriëntatie en kwaliteit van regulatie van docenten in een duale opleiding [The development of student teachers’ learning orientations and quality of regulation in a dual learning programme]. Paper presented at the Onderwijs Research Dagen 2009, Leuven, Belgium. Endedijk, M., Meijer, P., Vermunt, J., & Brekelmans, M. (2009). The development of student teachers’ quality of self-regulated learning and learning orientations in a dual learning programme. Paper presented at the ISATT 2009 Conference, Rovaniemi, Finland. Endedijk, M., Meijer, P., Vermunt, J., & Brekelmans, M. (2009). The development of student teachers’ quality of self-regulated learning in a dual learning programme. Paper presented at the 13th Biennial Conference EARLI 2009, Amsterdam, The Netherlands. Endedijk, M. D., Vermunt, J. D., Brekelmans, M., & Verloop, N. (2008). The quality of student teachers' self-regulated learning in innovative learning programs. Paper presented at the AERA 2008, New York.

203

Endedijk, M. D., Brekelmans, M., Vermunt, J., & Verloop, N. (2008). Individuele verschillen tussen docenten in opleiding in kwaliteit van zelfregulerend leren [Individual differences in student teachers’ quality of self-regulated learning]. Paper presented at the Onderwijs Research Dagen 2008, Eindhoven, The Netherlands. Endedijk, M. D., Brekelmans, M., Vermunt, J., & Verloop, N. (2008). The quality of student teachers' self-regulated learning in a dual learning environment. Paper presented at the 3rd biennial meeting of the EARLI Special Interest Group 16 Metacognition, Ioaninna, Greece. Endedijk, M. D., Brekelmans, M., Vermunt, J., & Verloop, N. (2008). The quality of student teachers’ regulation of learning in a dual learning environment. Paper presented at the The 4th EARLI SIG 14 Learning and Professional Development Conference, Jyväskylä, Finland. Endedijk, M. D., Brekelmans, M., Vermunt, J., & Verloop, N. (2008). Towards better understanding of the concept of self-regulated learning in teacher education. Paper presented at the ICO Winterschool, Jyväskylä, Finland. Endedijk, M. D., Vermunt, J., Brekelmans, M., & Verloop, N. (2008). The quality of student teachers' self-regulated learning in relation to their motivation to enter teacher education. Paper presented at the Motivation in Action. The 11th International conference on motivation, Turku, Finland. Endedijk, M. D., Vermunt, J., Den Brok, P., Brekelmans, M., & Verloop, N. (2007). The quality of student teachers' self-regulated learning in a dual learning environment. Paper presented at the 12th European Conference for Research on Learning and Instruction, Budapest, Hungary. Endedijk, M. D., Brekelmans, M., Vermunt, J., Den Brok, P., & Verloop, N. (2007). An exploratory study to student teachers' regulation activities in a dual learning environment. Paper presented at the The 10th conference of Junior Researchers of EARLI, Budapest. Endedijk, M. D., Vermunt, J. D., Brekelmans, M., Den Brok, P., & Verloop, N. (2006). Measuring self-regulation in complex learning environments. Paper presented at the ICO Toogdag, Amsterdam.

Posters Endedijk, M., Vermunt, J., Brekelmans, M., Den Brok, P., & Verloop, N. (2006). Zelfregulatie van leerprocessen van docenten in opleiding in nieuwe opleidingsarrangementen [Self-regulation of learning processes of student

204

Publications

teachers in new training arrangements]. Poster presented at the Onderwijs Research Dagen 2006. Endedijk, M., Vermunt, J., Brekelmans, M., Den Brok, P., & Verloop, N. (2006). Self-regulation of learning processes of student teachers in new training arrangements. Poster presented at the 9th JURE conference of EARLI 2006.

Other publications Endedijk, M. & Vermunt, J. (2008). The learning portfolio as a tool for stimulating reflection by student teachers [Bespreking van het boek The learning portfolio as a tool for stimulating reflection by student teachers]. Pedagogische Studiën, 85(1), 52-55 Endedijk, M.D. & Laroes, E. (2008). Rapport IVLOS alumni-enquête 2007. UU IVLOS.

205

Curriculum Vitae Maaike Endedijk was born in Westervoort, the Netherlands on April 30th 1982. She completed her secondary schooling in 1999 at the Corderius College in Amersfoort. From 1999 till 2001 she studied Psychology at the University of Groningen, for which she received the certificate of the first year examination. Next, she started to study educational science in which she graduated cum laude in 2004. She specialized in training and learning in professional organisations and in methods of educational research. As part of her master degree programme, she worked as a researcher at the University Centre for Learning & Teaching (UOCG, University of Groningen) to study student teachers’ professional development. In 2005, Maaike started as a PhD student at IVLOS Institute of Education, Utrecht University. Her research focused on student teachers’ selfregulated learning, which she presented at both national (ORD) and international conferences (EARLI, AERA, ISATT, JURE). Maaike received the award for the best poster presentation of the 9th JURE conference of EARLI 2006 and the award for the best paper presentation of the 10th Pre-Conference of Junior Researchers of EARLI 2007. As part of her PhD education, she followed masterclasses on teaching and teacher education, reflection on educational research, motivation and learning, and qualitative analysis. Furthermore, she visited three universities in Finland to discuss her research. During her PhD-project, Maaike worked as a junior teacher educator. Additionally, she was one of the organizers of the JURE 2008 conference. Currently, Maaike is working as a teacher educator and educational researcher at IVLOS Institute of Education. The focus of her research is on teaching and teacher education.

207

Organiseren

Meevoelen

Praktijkervaring

Steun

Samenwerking Reizen

Commentaar

Cursussen

Ideeën

Liefde

Ruimte

Inspiratie

Etentjes

Wijsheid

Concert

Kennis

Skypen

Telefoontjes

Creativiteit Logeren

Afleiding Collegialiteit

Muziek

Ontwikkeling

Adviezen

Belangstelling Dansen

Roddelen

Leren

Bewondering

Thee Voorbeelden

Vertrouwen Discussies

Wandelen

Literatuur

Ondersteuning

Gedachten Contact

Skiën

Begrip

Enthousiasme

Overleg

Blijdschap

Samen

Meedenken

Fietsen

Luisteren

Bijdrage

Lunchen

Heidagen

Snoeptrommel

Inzet

Feedback

Vertaling

Vliegen

Gesprekken

Delen

Verwondering

Suggesties

Begeleiding

Bedankt Inzichten

Lachen

Gutenberg

Motivatie

Support

Congressen Chocolade Vragen

Meedoen

Koffie

Onderwijs

Tournees

Kritiek

Passie

Borrel

Dankwoord Het schrijven van een proefschrift over zelfregulerend leren betekent niet dat de totstandkoming hiervan ook een geheel zelfregulerend proces is geweest. Dankbaar heb ik gebruik gemaakt van de sturing van anderen om mijn eigen weg te vinden. En daarom wil ik vooral jou bedanken. Jij, die ik niet bij name noem, omdat jij met zo velen bent. Omdat jij, met die ene opmerking, dat alleszeggende praktijkvoorbeeld, die inspirerende gedachten, die afleiding gevende roddel, dat ene kopje koffie, die veelzeggende blik, die prachtige muziek, dat gesprek op de gang, dat lange telefoontje, door samen te eten, te fietsen, te wandelen, te dansen, of misschien door er gewoon precies op dat moment te zijn - omdat jij mij, zonder dat je je er misschien van bewust was, dan weer net dat ene zetje gaf, waardoor een hobbeltje nooit hoefde uit te groeien tot een berg. Een speciaal woord van dank gaat uit naar alle dio’s [voor jullie enthousiaste medewerking aan mijn onderzoek], mijn begeleiders Jan, Mieke, Nico, Perry en Paulien [ik heb mij vereerd gevoeld met jullie rijkdom aan ervaring, kennis, ideeën, kritieken en enthousiasme, die al die jaren voor mij een onuitputtelijke voedingsbodem hebben gevormd], IVLOS onderzoekers [voor het gevoel van samen, onderzoek doe je nooit alleen!], IVLOS Lerarenopleiders [voor het meedenken, meewerken, mogen meekijken en meemaken], IVLOS collega’s [voor het warme thuisgevoel], ICO-, VOR-, en EARLI-collega’s [omdat onderzoek uitwisselen altijd eindigt in een feest], meiden van ALARM [voor al die psychologische inzichten], Kamerorkest Pulcinella [omdat in de muziek een dissonantie niet altijd tot een oplossing hoeft te komen - heerlijk], Papa, Mama, Hinke, Joanne [omdat we altijd onze eigen weg hebben mogen volgen, maar toch ontzettend dicht bij elkaar zijn gebleven] Larike [omdat ik door jouw komst mijn onderzoek kon afronden], Nienke [samen begonnen, samen geëindigd: ties with potential!], Miranda […..], Wietze […..].

209

skills months

on-line

making

giving

using

stimulate

calculated

passive goals changes

focus lack

really

content

relevant

nowadays

introduction turned semester

dimensions

self-regulation

regarded

random planning weeks characterised

works

trying

domain

standard

investigatin

simultaneously

the netherlands task

multiplelarg

externalone-yearresults

preprints

reliability items

beliefs

instruments values

approac urge stable psychologist strateg

its components

srl

almost

prospective different report obtain categorised outcome

intentional

below

levels

paid

rel

integrated association particularly analy lifelong self-reports meeting called aspects complex bec preoccupied insight thinking inventories

metacognitive internal information designed activities compared something spontaneous indicatio active contempora comparable related combination male journal perspec represents makes opportunities motivation iterative

per

interview

high-scoring

suitable

identified wanted

dissertation changing thesis

formulation

inv

clearly

paths

abbreviations

fre

evaluation defin students competences reflection dual specific otherwise underlying participants coded

actual

quantitative

expectations

actually

contribute observing

empirical schools

usability

coding

answers notions direction

asked

instructed

university observation

theory

methodological

issues

prepared

indicators

principal

cognitive

internship

completely

engaging quitted

caused

diaries

review

behaviour

average

meaning

characteristics

beginner

options

yourself

reflecting accepted

environment improve

recognised

linear cases limited

dynamic

conference

ins understanding available

teacher further preparation

independent

happened

directing

collab

freq

reproduct orchestration

deviation started improvement

testing separated planned description definitions

lacking

a

phase combined co suggestions

correspondence

assumption

pedagogy

categorisation

discriminating

unintentional extra school-based post-graduate upper relationships shortage

off-line

intraclass

separates

hig developed psychology away multiple measuring

metacognition examples received

ann

cognition

promote challenge

conducted

fitted

types

interpreted

contextual

unplanned effec colleagues responsibility di prog

succeed situation constructs

longitudinal qualities working things

education

mca

alternative

perform

measurement looki insights approache deliberate

qualitative stages

participate

dis

intercept exists peers consistency

closed

fourteen

satisfa

procedure

distinction examining myself learner portfoliodordrecht experienced steer

retriev

stimulated

requires

typical

c