Professional Doctorates and Careers: the ... - Wiley Online Library

4 downloads 3717 Views 159KB Size Report
European doctoral education is undergoing a very rigorous analysis and ... for researchers who wish to have a university career and that public and private.
European Journal of Education, Vol. 47, No. 1, 2012

Professional Doctorates and Careers: the Spanish case1 ejed_1514

153..171

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez Introduction Why do people embark on doctoral studies? What returns do they expect? Should a country increase the number of PhD graduates? According to Kehm (2007), European doctoral education is undergoing a very rigorous analysis and adapting to the new education policies. In this sense, the 1999 Bologna Declaration or the 2000 Lisbon Strategy have prompted decisive changes in how third-cycle education is regarded. The current definition of doctoral education is different from its initial concept (Noble, 1994; Clark, 1995). At the beginning, in order to obtain a PhD, students had to prove they had the skills to teach at a university. However, welfare state expansion and the development of public university education as from the 1960s meant a change in how the university’s ultimate target was perceived. It became a place not only for education, but also for research. The most immediate consequence was the emergence of practical learning in research (Ben-David, 1992; Jamieson & Naidoo, 2007). Research learning became a need for students who wished to become university teachers or researchers. Hence, the University is not only pressured by socio-economic changes, but also by its own organisational ones which have triggered a transformation in the meaning of the PhD title and its assessment in the labour market. So, as Enders (2004) and Kehm (2007) point out, the labour market’s request for doctors has increased remarkably for several reasons. Among these are the fact that being a PhD graduate has become a ‘must’ for researchers who wish to have a university career and that public and private firms need to recruit more people with research experience. The Spanish University is aware of these changes. Perotti (2007) says that Spain is a unique case in the rapidity with which they have been implemented. However, he considers that in order to clearly understand this process, both socio-economic (university admission rules, overcrowded classrooms) and academic changes must be analysed individually. It is particularly important to assess the contents of the programmes and students’ opinions to make the right decisions and adapt them to the changing European university. The first and second university education cycles have been analysed in Spain (Sanchez, 1996; Mora et al., 2000; Mora & Vidal, 2000), but not doctoral education because of a lack of databases. Buela-Casals and Castro (2008) studied its development from a quantitative perspective, generating lists of universities according to the number of high-quality PhDs. As the rest of Europe also lacks information, the EU backed surveys on this cycle by passing Regulation 753/2004 on science and technology which defines the framework to generate statistics about PhD graduate workers. The National Statistics Institute (INE) carried out the ‘2006 Survey on Human Resources in Science and Technology’ in 2008, which was an exhaustive study on doctors who obtained their degree between 1990 and 2006 at any Spanish university. This article outlines their characteristics. In particular, it © 2012 Blackwell Publishing Ltd., 9600 Garsington Road, Oxford OX4 2DQ, UK and 350 Main Street, Malden, MA 02148, USA.

154 European Journal of Education, Part II analyses whether there is a tendency amongst PhD graduates to change their university career when offered private or public firm opportunities. This information proves fundamental in the framework of the changes undergone by Spanish and European universities in both the structure of the PhD programmes and their generation. Changes in the PhD Model Many papers have studied the recent changes in universities (Abbot, 2001; Naidoo, 2003; Naidoo & Jamieson, 2005). According to these, universities have changed from training a selected group of people to having to face massification problems and from being knowledge-generating institutions to becoming institutions that train people to deal with their daily work problems (Gibbons et al., 1994). As Jamieson and Naidoo (2007) stress, it would be surprising that doctoral education were to be unaware of these changes. The excess supply of people with higher education has led to its ‘being devalued’ in the labour market and has created the need to incorporate an extra ‘requirement’ in master and doctoral education where students have command of some knowledge in the former and innovate in the area of knowledge in the latter. Enders (2004) and Kehm (2007) consider that worldwide demand for doctoral education increased by 30% in the 1990s. This forced a change in teaching methods from a learning process supervised by tutors to a mixed model where training is ensured by the institution and supervision is shared with the tutors (Kehm, 2007). Such massification also caused problems when assessing candidates’ research skills and their interest in an academic career. In this case, the solutions proposed by the system are either to design a selection exam or carry out a selection process during doctoral education. Such changes have not only affected the way the PhD works, but also its content. National governments show growing interest in university research funds’ return because the university is regarded as an institution whose targets are closely related to those of firms, where efficiency and economies of scale become more important and students are considered as consumers. All this within an international framework where there is great competition to have the most qualified labour force available (Brooks & Heiland, 2007). National and supranational institutions (e.g. the European Commission) have established a definition of the PhD according to society’s needs to obtain some return and a more immediate application of the knowledge acquired at university. They have encouraged more practical university research which is also more closely related to the non-university world (Häyrinen-Alestalo & Peltola, 2006). The remarkable growth in the number of doctoral candidates and in the variety of research fields in Europe and the US has made the professional university career less accessible for students who are looking for a job outside the university. In this context, traditional doctoral education directed towards university teaching proves to be insufficient (Crosier et al., 2007) This idea is also developed by Jamieson and Naidoo (2007) who find the emergence of two doctoral models because of labour market pressures. The first is the American model developed by more than 30 US universities and supported by their government. It would be a variation of the traditional model in which, in order to obtain a PhD, a piece of research must be defended in front of a committee. This model focuses on students’ learning process so that the lack of knowledge in the research area might be rectified and students should be given © 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

155

knowledge in research theories and methodologies beyond their specialised area of studies.The second model is related to the growing number of professionals taking up doctoral studies, especially in the US, Australia and the UK. In this case, besides carrying out research and defending a thesis in front of a committee, candidates carry out research to solve a problem that is considered vital for a given profession. Future Doctoral Education Guidance Both the 1999 Bologna Declaration and the 2000 Lisbon Strategy considered as strategic the supply of the highest qualified human resources to reach the greatest economic and technological growth. However, at the European level, there is great concern about the poor number of researchers. The causes could be found in a decreasing interest among students to join certain science research fields and EU difficulties to retain the most brilliant researchers (Moguérou, 2005). To show the importance of this problem for the EU’s economic development, the European University Association in its 2008 Trends V report stressed that the EU was concerned about PhD employment. Hence, it supported a model that fostered the relation between the university and firms (public and private) in such a way that a university career was not presented as the only option for PhD holders. To sum up, two tendencies can be distinguished in the identification of reform targets and the analysis of the tools and models used for their implementation. Doctoral education and research training cannot be considered to be simply devoted to an unselfish search of knowledge. The creation of knowledge has become a basic strategic resource for developed economies and is beginning to be treated as a good. As it is considered such an important resource, it cannot be left in the hands of teachers and departments and it becomes a component when formulating national or even supranational policies. But the remarkable growth in the number of PhD graduates will transform looking for employment outside the academic institutions into a challenge. Such employment is necessary for that qualified labour force to boost economic growth and innovation. However, for such jobs, research training directed towards academic teaching is not considered enough, so doctoral education must change (Kehm, 2007). To face the changes in doctoral programmes as a result of their being oriented towards the labour market outside the university, one must analyse which parts of the current model must be reviewed and which ones work regarding the final targets. Hence, we shall now describe the content of the survey carried out by the INE on PhD graduates in Spanish universities in order to obtain the most accurate image of the success of their current doctoral education programmes. What is the Doctors’ Status in Spain? The 2006 Survey on Human Resources in Science and Technology is an exhaustive study of doctors who obtained their title between 1990 and 2006 at any public or private Spanish university2. The total number of selected individuals is 17,000, the final total sample population being 12,625. It took the year 2006 as a basic reference, although other periods were included according to the theme areas into which the survey was divided. Finally, the content was divided into different areas, providing information about personal characteristics, areas of research, labour status, international, national and sector mobility, scientific output, subjective assessment of the decision to follow research training, and wages. © 2012 Blackwell Publishing Ltd.

156 European Journal of Education, Part II Regarding personal characteristics, 45.2% were women and the sample average age was 41, the most frequent being 38 (749 cases, i.e. 6% of the sample). This figure should be considered low, given that the target group corresponds to those with the highest training profile in this country. Doctors’ distribution by areas of knowledge reveals that three gathered the most remarks: natural sciences (29.2%), health (22.6%) and social sciences (20.8%). Far behind, we see humanities (14%), engineering and technology (9.6%), and agricultural science (4%). Concerning the year the titles were obtained, the fact that doctor databases are new determines the results, as 54% obtained the title after the year 2000. Despite this, there has been a growth in the number of new doctors since 1990.This follows the observed international behaviour that changed trends in 2003 because of students’ decreasing interest in joining the research field of some science areas (maybe because of the greater job opportunities offered to bachelors by the labour markets at that time) and the EU’s difficulty to retain the most outstanding researchers (Moguérou, 2005). In the case of Spain, the most remarkable drops between 2003 and 2006 were in social sciences and humanities (47% and 44% respectively). Once the PhD title is obtained, entering the labour market does not seem a problem, since, by December 31st 2006, the activity rate was 96.5%, well above that for the whole Spanish labour market. The unemployed represented 2% and the inactive 1.5%. Almost half the PhD graduates (48.3%) work in higher education institutions, with health and hospital-related activities being the second most frequent activity (16.4%). 10% were in natural science and technical research and development activities. 44% worked as university teachers and 18.3% as doctors and related professions (except nursing). The remaining professions are heterogeneously distributed, always below 10%. Since 44% stated that they were working in higher education institutions and 36% in the Public Administration, it can be said that almost 80% belonged to the public sector; 14.8% were in private firms, whilst the others worked for non-profit institutions. These data are consistent with the European labour market tendency to require research-trained individuals (in both public and private firms), as more than half the PhD graduates do not work in education institutions. But despite the increasing trend towards jobs outside the university, being a university teacher is still the predominant activity. There are differences in the area of knowledge. Teaching is the most frequent activity, but it is 20% higher in the case of humanities and social sciences. 42.5% of doctors of science are teachers, whereas they are 62.7% in the humanities and social sciences. Hence, there is a tendency among the latter to choose a traditional university career, whilst doctors in science tend to follow a career outside the university. Moreover, there is an outstanding educational vocation among doctors of humanities and social sciences because, if all the professional options in the education field are summed up, they represent more than 78%. But the distribution of doctors of science among the professional options is much more dispersed, as more than 34% are found in research institutions and in health and hospitalrelated activities. Concerning labour relations, most doctors (94%) work full-time with a permanent contract (72%). However, temporary work rates do not differ from those observed for the Spanish labour market as a whole. This has become particularly worrying, since hardly 12% belong to the private sector. This means that the

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

157

problem is located in the public sector, mainly in higher education where half the part-time workers are to be found. Concerning earning levels, as we can see in Table I, there are differences in relation to those that the labour market establishes for research training where the most favoured are those working in health areas. At the opposite end, we find humanities, agricultural sciences and natural sciences. Engineering and technology and social sciences are in the middle. Table I. Earnings according to area of knowledge (percentage of workers for each interval) Interval Less than 10,000€ From 10,001€ to 20,000€ From 20,001€ to 30,000€ From 30,001€ to 35,000€ From 35,001€ to 40,000€ From 40,001€ to 45,000€ From 45,001€ to 50,000€ More than 50,000€

Natural Sciences

Engineering and technology

Health

Agricultural Sciences

Social Sciences

Humanities

2.78 13.96

1.69 9.81

1.18 7.68

1.86 14.23

2.73 11.79

6.11 16.53

30.90

23.43

13.34

29.90

24.28

30.21

18.13

17.51

11.26

22.47

16.63

16.65

13.62

17.01

12.91

12.99

16.28

11.20

9.90

10.91

11.51

9.48

11.79

9.62

5.31

7.53

13.31

4.74

6.83

5.87

5.39

12.10

28.80

4.33

9.68

3.81

To sum up, these data present some interesting features. First, there is an increasing tendency to train new doctors, although it slowed down in the last years of the survey. This reflects the labour market’s need for a highly qualified workforce where both men and women are equally incorporated. Second, the most common professional choice is university teacher, following the academic tradition for this type of training. However, universities are no longer the most popular option for doctors of science, as more than half work in non-university institutions. This may go hand in hand with firms’ growing demand for doctors. Finally, there seems to be a difference in wage level depending on the area of knowledge and this difference favours science studies. Based on these conclusions, we shall develop an econometric model to analyse the factors that determine the choice of professional activity and show that they can be explained by an Economic Theory. This model will also take into account the fact that choice may vary depending on the area of knowledge. Factors Determining the Career of PhD Graduates The factors determining people’s decisions when facing various options can be analysed by different discrete choice models. The advantage of these models vs. traditional econometrics is that the former allows for the modelling of qualitative variables by using discrete variable techniques. Here, it was decided to use a multiple choice model that can be applied when the endogenous variable is a discrete variable with alternative answers. © 2012 Blackwell Publishing Ltd.

158 European Journal of Education, Part II In this article, selection refers to the professional alternatives chosen by doctors. It is mainly conditioned by the area of specialisation, as training is presented as the fundamental factor of a person’s human capital and will therefore be a key factor when choosing a career. The data show that individuals chose among 27 professional activities according to their classification by the ISCO-88 National Classification of Occupations. As 44% of the interviewees work as university teachers, it was decided to divide professional selections into two groups: university teacher and other professional activities. In relation to field, according to Stephan (1996), sciences attract the attention of economists by being a source of economic growth and a convenient labour market for the economic study. However, an in-depth analysis of PhDs’ labour market requires including all fields of knowledge, so the sample was divided into doctors of sciences and doctors of humanities and social sciences3. The variables influencing the selection of a professional career can be divided into three groups: personal characteristics, training and research, and job characteristics. Earnings are found in this last group. Since it is not possible to determine whether earnings are the consequence of professional choice or whether the professional choice was made based on the expected earnings, a Mincerian wage equation was estimated to avoid the likelihood of endogeneity and followed Labour Economy traditional specifications LnW = Xβ + u where LnW stands for wage logarithm, X is the characteristics vector, b is a parameter vector and u the random errors distributed independently in a normal way with 0 average and σ u2 (0, σ ) variance. The estimated salaries have recently been entered into the multinomial logit. Results Descriptive Statistics Table II gives the statistics for the variables used in the estimations (see also Table A1 in the Appendix). As can be seen, there are substantial differences in their values, as they take into account both doctors’ area of knowledge and their professional activity. Concerning personal characteristics, because women joined universities much later, the percentage of men is higher in both fields of knowledge, and even higher in sciences (this is normal given the traditional male profile in such activities)4. Regarding age, the average is higher in the case of humanities and social sciences (7.6%).This may be the result of a recent development of some science fields (especially new technologies) which has created the need to train more doctors to meet the demand for qualified workforce. Concerning professional activity, for humanities and social sciences, the age is higher for PhD holders who are not teachers. It is much higher in the case of humanities and social sciences because of the large percentage of men. Regarding training and the time people need to obtain their degree, the data state that 18% extra time is required in the humanities because of the different characteristics of the doctoral studies depending on the field of knowledge. They are always longer for those who also work.This may be related to Spanish students’ difficulty in combining research training and work in a firm. Post-doctoral studies are another interesting element, as they are complementary to PhD training. Data show that a high percentage (17%) has taken postdoctoral studies in both sciences and humanities, whilst humanities and social sciences figures are very low for those working outside the university. © 2012 Blackwell Publishing Ltd.

© 2012 Blackwell Publishing Ltd.

Job characteristics Public sector Worked hours Full time Permanent contract

0.78 41.10 0.95 0.71

5.47 0.17 0.30 1.33 5.72

Training and research: PhD length Post-doc studies International mobility Published books Published papers

0.70 0.04 0.26

1.29

Married Other Single

0.56 39.75

People under his responsibility

Personal characteristics Male Age Marital Status

Mean

0.41 8.77 0.21 0.45

2.94 0.37 0.46 2.78 7.34

1.31

0.46 0.20 0.44

0.50 7.38

St dev.

Science

0.87 37.17 0.91 0.75

6.48 0.17 0.26 2.80 5.65

1.14

0.66 0.07 0.27

0.52 42.77

Mean

0.33 10.20 0.29 0.43

337 0.37 0.44 3.95 9.21

1.24

0.47 0.26 0.44

0.50 8.57

St dev.

Humanities and Social Sciences

0.99 40.72 0.96 0.75

5.38 0.18 0.36 1.79 7.65

1.28

0.72 0.05 0.24

0.62 39.69

Mean

0.09 8.33 0.20 0.43

2.59 0.38 0.48 2.99 8.12

1.22

0.45 0.21 0.43

0.49 6.76

St dev.

University teacher

0.66 41.32 0.95 0.69

5.54 0.16 0.26 1.08 4.60

1.30

0.69 0.04 0.27

0.53 39.89

Mean

0.47 9.02 0.22 0.46

3.13 0.37 044 2.65 6.65

1,36

0,46 0,20 0,44

0,50 7,57

St dev.

Other professional activity

Science

0.98 37.74 0.91 0.71

6.23 0.24 0.32 3.50 6.57

1.13

0.66 0.08 0.26

0.49 41.80

0.13 10.75 0.28 0.45

3.15 0.42 0.47 4.24 6.27

1.20

0.47 0.26 0.44

0.50 7.81

St dev.

0.70 36.29 0.89 0.81

6.81 0.06 0.15 1.78 4.01

1.16

0.66 0.07 0.27

0.59 43.80

Mean

0.46 9.23 0.31 0.39

3.46 0.24 0.36 3.29 7.43

1.30

0.47 0.25 0.44

0.49 8.69

St dev.

Other professional activity

Humanities and Social sciences University teacher Mean

Table II. Descriptive statistics of variables included in the estimation

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez 159

Less than 10,000€ From 10,001€ to 20,000€ From 20,001€ to 30,000€ From 30,001€ to 35,000€ From 35,001€ to 40,000€ From 40,001€ to 45,000€ From 45,001€ to 50,000€ More than 50,000€

Post-doc Doctor Graduate Undergraduate Professional training

0.02 0.11 0.23 0.16 0.14 0.11 0.08 0.14

0.09 0.33 0.52 0.03 0.02

0.59 0.22 0.19

8,698

0.14 0.31 0.42 0.37 0.35 0.31 0.28 0.35

0.29 0.47 0.50 0.18 0.15

0.49 0.42 0.39

St dev.

0.05 0.15 0.28 0.16 0.13 0.10 0.06 0.06

0.05 0.41 0.47 0.04 0.03

0.64 0.21 0.15

Mean

3,495

0.21 0.36 0.45 0.37 0.34 0.30 0.24 0.25

0.21 0.49 0.50 0.21 0.17

0.48 0.41 0.36

St dev.

Humanities and Social Sciences

0.01 0.09 0.24 0.20 0.18 0.14 0.08 0.06

0.11 0.60 0.25 0.03 0.00

0.81 0.15 0.04

0.11 0.28 0.43 0.40 0.38 0.34 0.27 0.24

0.31 0.49 0.44 0.18 0.02

0.39 0.36 0.20

St dev.

3,244

Mean

University teacher

Note: Unemployed and inactive doctors are not included because information on earnings is not available.

No. of observations

Earnings

Minimum training level

Relation between job and doctoral studies High Normal Low

Mean

Science

Table II. Continued

0.02 0.12 0.23 0.14 0.12 0.09 0.09 0.19

0.08 0.18 0.67 0.03 0.04

0.45 0.27 0.28

Mean

5,454

0.15 0.33 0.42 0.34 0.33 0.29 0.28 0.39

0.28 0.38 0.47 0.17 0.18

0.50 0.44 0.45

St dev.

Other professional activity

Science

0.04 0.14 0.26 0.16 0.14 0.12 0.07 0.06

0.06 0.63 0.29 0.02 0.00

0.84 0.13 0.03

0.19 0.35 0.44 0.37 0.35 0.32 0.25 0.24

0.24 0.48 0.45 0.12 0.04

0.37 0.34 0.17

St dev.

2,118

Mean

University teacher

0.06 0.16 0.30 0.16 0.12 0.08 0.05 0.07

0.02 0.07 0.75 0.09 0.07

0.34 0.33 0.33

Mean

1,377

0.24 0.36 0.46 0.37 0.32 0.28 0.22 0.26

0.14 0.25 0.43 0.29 0.26

0.47 0.47 0.47

St dev.

Other professional activity

Humanities and Social sciences

160 European Journal of Education, Part II

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

161

International mobility is mainly found in the science fields (four percentage points higher) and, regardless of the knowledge field, among university teachers, as the research activities are frequently related to international research groups and stays in international universities. Finally, the contents of both knowledge areas determine the quantification of the scientific outputs. So, in the fields of humanities and social sciences, the average book publication rate is double that of sciences, whilst the average paper publication is higher in the case of sciences. Scientific output is higher for university teachers, as it is determinant for both their professional careers and their earnings.5. Regarding job characteristics, most doctors work for the public administration (78% in the case of sciences and 87% in the case of humanities and social sciences), which indicates the mismatch between research training and the workforce demand among firms in Spain. In relation to professional activity, percentages are close to 100% in the case of university teachers, which shows the lack of interest of Spanish doctors for the private university. There are remarkable differences regarding the match between doctoral studies and the job, depending on the activity. As for the field of knowledge, doctors of humanities and social sciences consider this relationship clearer than doctors of sciences, while, regarding professional activity, university teachers consider this relationship clearer than non teachers. These figures reveal that doctoral education still focuses on the university world, so when they are outside this world, workers find it difficult to put their knowledge into practice, either because they do not find a job that is related to their area of knowledge or because their training level is not required. In this sense, the variable that tries to measure over-education (the minimum level of studies required for a job) detects the lack of a match between their studies and job requirements, especially for doctors who are not university teachers. Around 60% of university teachers consider the PhD as the minimum necessary qualification. However, this percentage should not be considered quite high since it is the minimum requirement for a university career. As for those in other activities, 91% of doctors of humanities and social sciences do not consider the PhD degree to be necessary for their jobs. In the case of doctors of science, this percentage is 74%. Both percentages prove a great lack of coordination between the contents of the doctorate courses and the knowledge required by firms. Finally, Table II shows that the percentage of people earning less than 30,000€ a year is lower in science than in the humanities and social sciences. In relation to professional activity, the percentage of doctors of sciences who work outside the university with the highest earnings is higher than that of those who work at the university. Meanwhile, the percentage of doctors of humanities and social sciences who work outside the university with the lowest earnings is higher than the percentage of those who work at the university, i.e. doctors of humanities and social sciences earn less than those of sciences, regardless of their professional activity. Working outside the university seems to be positive in relation to earnings for doctors in sciences and negative for those in humanities and social sciences. Results According to the econometric model, the first step is the estimation of the wages equation in order to correct wages endogeneity. Table AII of the Appendix shows © 2012 Blackwell Publishing Ltd.

162 European Journal of Education, Part II the expected results following the human capital theory and the descriptive statistics mentioned in the previous section. The estimated wages are included in the multinominal logit independent variable group. Results are shown in Table III and marginal effects in Table IV. Following the results in the latter table, it can be observed that, when increasing the average age of doctors, the likelihood of choosing any science area is reduced, while the chances of selecting a humanities and social area increase. This seems logical, as in the last decades there has been a very great development in the science area which has caused a growing demand for professionals trained in this area. In relation to sex, it does not have a clear effect on either the selection of the area of knowledge or the professional career. In relation to training and research, doctoral education procedures and methodology are quite different for the science and for the humanities and social areas. This is because any delay to obtain the PhD in science is negatively valued (even though it is not statistically significant for university teachers), while it is positively valued in the humanities and social areas for any professional activity. As for doctors of humanities and social sciences, to take post-doc education does not exert a positive effect on the chances of developing any type of professional career. For doctors of sciences, a positive effect is only observed when choosing a job outside the university. Both results indicate the insufficient development of this research option and its little value in the labour market. Scientific output has different effects depending on the area of knowledge. Publishing books has a positive effect on the development of any professional activity in the case of doctors of humanities and social sciences and obviously it is greater among university teachers. As for doctors of science, the negative effect is higher among non-university teachers but is not relevant for university teachers. Publishing papers is much more related to the area of sciences, as it has a positive effect on any job (greater in the case of university teachers) but has a negative effect for doctors of humanities and social sciences. Finally, international mobility does not affect the development of a professional career. Regarding job characteristics, working for the public sector favours the likelihood of becoming a university teacher. But the negative effect of a full-time job and of the worked hours on the chances of developing a university career will show that working at the university is compatible with other professional activities outside the university. These should be regarded as complementary to the main activity (university teaching and research) and as the natural and expected link between the university and the scientific and business worlds. Among university non-teacher doctors, both variables have a positive effect (greater in the case of doctors of science), although it is negative in the case of doctors in humanities and social areas. Maybe, in this case, the different working methods identifying labour activity are gathered depending on whether the worker is a doctor of science or humanities and social sciences. The match between training received and the job also influences professional activity. So, to state a clear relation strengthens the possibility of choosing to become a university teacher, which is more noticeable among doctors in science. Similarly, the higher the minimum level of required studies to carry out the job, the higher the chances of becoming a university teacher. This effect is also greater © 2012 Blackwell Publishing Ltd.

© 2012 Blackwell Publishing Ltd.

-6.02 -5.36 5.68 3.38 3.40 -2.69 3.34 -11.72 3.57 3.15 0.24 13.30 -5.06 -9.95 21.85 9.57 16.04 5.33 6.54 3.73 5.26 -10.96

-4.811

-0.026 0.330 0.256 0.500 -0.072

0.035 -0.921 0.042 0.013 0.017

1.012 -0.702 -0.037 4.281 1.101 1.737 3.866 4.715 2.689 3.851

-0.906

-1.431

0.823 -0.823 -0.086 3.629 1.356 2.309 2.491 3.918 1.929 2.169

0.098 -0.569 0.205 -0.018 0.100

0.026 -0.136 0.164 0.614 -0.178

-2.973

Coefficient

-15.54

9.50 -5.81 -19.53 19.21 8.51 15.35 4.08 6.52 3.22 3.42

9.04 -6.61 18.37 -3.55 1.28

5.12 -2.07 1.91 3.94 -5.64

-4.23

Standard error

University teacher. Humanities and Social areas (Choice = 3)

Dependent variable basic option is: Other professional activity. Sciences area (Choice = 2). The reference variables are: Single, Relation job/PhD low and Minimum training level: professional training.

No. of observations: 12,193 c2 = 8,689.42 Prob > c2 = 0.00 Pseudo R2 = 0,28

Constant Personal Characteristics Age Male Married Other marital status Dependent people Training and research PhD length Taking a post doctoral Published books Published papers International mobility Job characteristics Public sector Permanent contract Full time Worked hours Relation job-PhD high Relation job-PhD normal Minimum training level: post-doc Minimum training level: doctor Minimum training level: graduate Minimum training level: undergraduate Earnings Estimated wages

Standard error

Coefficient

University teacher. Sciences (Choice = 1)

Table III. Multinomial Logit

-1.460

0.750 -0.035 -0.084 0.291 0.227 0.084 -1.076 -1.107 -0.350 0.429

0.077 -0.511 0.144 -0.013 -0.079

0.060 0.417 -0.002 0.396 -0.166

2.389

Coefficient

-16.64

8.03 -0.26 -18.92 3.69 2.65 0.94 -4.15 -5.90 -2.38 2.28

7.82 -3.80 11.30 -2.13 -0.84

12.56 5.97 -0.03 2.53 -5.33

7.36

Standard error

Other professional activity. Humanities and Social areas (Choice = 4)

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez 163

3.44 5.81

-6.72

0.377

0.647

-0.084

42.73 12.38 -3.91 -4.02 14.19 6.29 5.05

0.324 0.119 -0.098 -0.002 0.201 0.143 0.638 4.96

1.16 -12.25 -0.64 4.69 0.17

0.002 -0.109 -0.001 0.003 0.002

0.619

-8.32 5.73 3.43 2.42 -1.03

-0.006 0.048 0.037 0.059 -0.004

Statistic z

17.70

-13.98

-0.545

0.288

-5.16

-0.400

-43.97 -15.53 5.56 19.94 -20.85 -11.80 -13.01

-0.448 -0.206 0.154 0.015 -0.333 -0.249 -0.557 -15.40

-7.56 10.99 -11.20 0.55 -0.34

-0.015 0.164 -0.027 0.0001 -0.005

-0.644

-2.15 -4.45 -2.76 -4.24 5.63

Statistic z

-0.002 -0.052 -0.041 -0.127 0.030

Dy/dx

Other professional activity. Sciences area (Choice = 4)

-12.01

-0.25

-0.019

-0.092

1.89

1.94

27.70 7.24 -4.25 -15.52 14.14 5.53 0.28

8.26 -4.70 16.65 -4.22 1.48

5.57 -4.26 1.40 2.85 -4.97

Statistic z

0.121

0.197

0.164 0.052 -0.079 -0.007 0.178 0.130 0.027

0.009 -0.033 0.019 -0.002 0.011

0.003 -0.027 0.011 0.053 -0.015

dy/dx

University teacher. Humanities and Social areas (Choice = 3)

The marginal effects were calculated as an average over every covariate average. The reference variables are: Single, Relation job/PhD low and Minimum training level: professional training.

Personal Characteristics Age Male Married Other marital status Dependent people Training and research PhD length Taking a post doctoral Published books Published papers International mobility Job characteristics Public sector Permanent contract Full time Worked hours Relation job-PhD high Relation job-PhD normal Minimum training level: post-doc Minimum training level: doctor Minimum training level: graduate Minimum training level: undergraduate Earnings Estimated wages

Dy/dx

University teacher. Sciences area (Choice = 1)

Table IV. Marginal effects

-0.112

-0.083

-0.098

-0.172

-0.040 0.036 0.023 -0.006 -0.046 -0.024 -0.107

0.005 -0.022 0.009 -0.001 -0.008

0.005 0.031 -0.007 0.016 -0.011

dy/dx

-12.15

-8.30

-5.87

-15.34

-5.46 5.32 2.55 -14.10 -6.01 -3.58 -17.21

6.00 -2.24 9.05 -2.29 -1.08

12.97 5.36 -0.91 1.07 -4.14

Statistic z

Other professional activity. Humanities and Social areas (Choice = 4)

164 European Journal of Education, Part II

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

165

among doctors of science. To sum up, both effects prove a greater suitability of doctoral education for the development of a university career, especially in the case of doctors of science. Finally, the effect of the expected wages penalises the selection of any professional career in the humanities and social sciences, i.e. the low wages cause it to be less likely to be chosen than the science option. Besides, this effect is also negative when selecting a university career, due to the lower wages of university teachers. A positive effect of the likelihood of developing a professional career outside the university can only be estimated for doctors of science. So, wages’ negative effect penalises choosing PhD studies in humanities and social sciences and a university career. However, according to Schomburg (2007), doctors present a more complex career orientation than other workers and so there are many variables, besides wages, that can be used to measure profitability and professional success (work stability, social recognition, contribution to welfare state, satisfaction of developing an innovating activity, etc.). This means that wages’ marginal effect reflects the importance of the wages expected when deciding on the training and the professional career to follow, as well as other non-monetary and significant characteristics for the individual in relation to the job position. In order to value the non-monetary characteristics, the survey includes questions about people’s satisfaction with some qualitative characteristics of the job (Table V). In general, when we take into account the area of knowledge, there is great satisfaction among the doctors in humanities and social sciences. Given the negative marginal effect of wages for the two options related to this knowledge area, it can be said that the expected low wages will not compensate for the satisfaction derived from the non-monetary characteristics of a job. On the other hand, if we take into account the professional activity, doctors who are university teachers show great satisfaction. As for the previous case, the group presents a negative marginal effect of wages, and so we may also conclude that the low wages expected in relation to other professional activities will not compensate for the positive ones derived from the satisfaction of doing their job. In the case of doctors of sciences who are not university teachers, if we analyse the professional choices, the positive marginal effect of wages indicates that the high expected wages will compensate for the less valued characteristics of the job such as stability, labour conditions, the contribution to society or social status. As an average, in the case of doctors of humanities and social sciences, the university teachers declare great job satisfaction. However, the negative marginal effect of the expected wages will not compensate for the positive characteristics of their jobs. Besides, doctors who carry out other activities have the worst job satisfaction assessment.Thus, wages’ marginal effect is negative and greater than in the case of university teachers. This indicates that not only do wages not compensate for a negative job assessment, but also, given that the level of expected wages is the lowest in the analysed groups, wages disincentive is the highest of all when choosing this option. Finally, if we compare the groups of teachers of sciences and humanities and social sciences, we see that they also present the negative marginal effect of the expected wages, which is less powerful for doctors of sciences, maybe because of their greater expected wages. If we compare the satisfaction of both groups, the © 2012 Blackwell Publishing Ltd.

166 European Journal of Education, Part II Table V. Satisfaction level in relation to some qualitative characteristics of the job (figures in percentages) Sciences Science

Humanities and Social sciences

Earnings High 15.0 13.2 Medium 52.8 53.5 Low 29.9 30.6 No 2.3 2.7 Job stability High 56.3 60.1 Medium 19.7 19.2 Low 16.1 14.6 No 7.9 6.1 Job conditions High 36.5 37.4 Medium 46.5 46.2 Low 15.3 14.3 No 1.6 2.2 Opportunities to be promoted High 13.0 13.0 Medium 37.4 37.4 Low 37.3 37.3 No 12.4 12.4 Level of responsibility High 54.4 52.8 Medium 37.8 38.8 Low 7.2 7.4 No 0.7 1.0 Intellectual challenge High 55.3 54.0 Medium 30.2 29.7 Low 11.8 12.8 No 2.7 3.5 Contribution to society High 56.8 58.5 Medium 36.1 35.5 Low 6.2 5.2 No 0.8 0.9 Social status High 20.8 22.4 Medium 63.9 61.5 Low 13.0 14.0 No 2.3 2.0 No. Observs.

8,698

3,495

Humanities and Social sciences

University teacher

Other professional activity

University teacher

Other professional activity

12.1 50.8 34.2 3.0

16.8 54.0 27.4 1.9

11.9 52.3 33.0 2.9

15.3 55.4 27.0 2.3

67.5 14.3 12.7 5.5

49.7 22.9 18.1 9.4

59.0 18.9 16.5 5.6

61.7 19.6 11.8 7.0

43.3 44.3 11.4 1.0

32.5 47.8 17.7 2.0

39.3 45.6 13.4 1.7

34.4 47.1 15.6 3.0

13.6 47.6 33.3 5.5

12.6 31.3 39.7 16.4

19.3 46.4 28.5 5.8

10.9 24.0 40.2 24.9

45.5 46.7 7.4 0.5

59.7 32.5 7.0 0.8

56.3 37.6 5.5 0.7

47.5 40.7 10.3 1.5

73.2 22.3 4.2 .3

44.6 34.9 16.4 4.2

69.0 25.0 5.4 .6

30.9 37.0 24.2 8.0

57.2 38.7 4.0 0.1

56.6 34.6 7.5 1.2

60.1 36.4 3.1 0.4

56.0 34.0 8.4 1.6

21.8 64.4 11.7 2.1

20.3 63.5 13.8 2.4

27.3 61.5 10.0 1.2

14.7 61.6 20.3 3.3

3,244

5,454

2,118

1,377

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

167

science teachers greatly value job stability, labour conditions and intellectual challenge, while humanities and social science teachers greatly value promotion, level of responsibility, contribution to society and social status. Conclusion This article analysed the determining factors for doctors when choosing a professional career by using the information provided by the ‘2006 Survey on Human Resources in Science and Technology’. Choosing a professional career is mainly influenced by doctors’ area of knowledge. Hence, in this article, the survey differentiated between science and humanities and social science doctors in order to prove whether the professional projection differs when it is technical training or social sciences-oriented. In relation to the professional alternatives, since most doctors work as university teachers, it was decided to divide professional choices into university teachers and other professional activity. To estimate the results, a multinomial logit model was applied where wages endogeneity was controlled by means of the estimation of a Mincerian equation. From the analysis of the descriptive data and the results of the estimations, several interesting conclusions could be useful when designing doctoral education course guidelines. As in the case of the rest of the nearby countries, a constant increase is observed in the graduation of new doctors caused by the Spanish public university’s need to expand continuously and by a firm sector (public and private) encouraged by the economic growth of our country in the last two decades which demanded a highly-qualified workforce. The area of knowledge with a weaker doctor contribution is humanities with hardly 30%. Besides, a decreasing relative weight throughout the analysed period is revealed. According to the survey data, doctors’ professional choice has undergone a remarkable change. Currently, less than half the working doctors work in higher education institutions. This means that public and private firms are beginning to absorb the growing numbers. However, behaviour is different according to the area of knowledge, as, even if the university keeps on being the commonest choice for doctors of humanities and social sciences, a job outside the university is now becoming the main option for doctors of sciences. Besides, the results of the estimations indicate that when choosing a professional career, people’s research skills and research training period are taken into account. As for the Spanish case, it is observed that the labour market does not yet value either post-doc education or international mobility. For every knowledge area, the professional career at a public firm is linked mainly to the university environment where it seems easier to make compatible both teaching and research activities according to the type of contract and the time devoted to the main labour activity. It is in this group that a better suitability between the training received and the professional activity developed is seen, which may determine the selection of a professional future. Finally, the expected wages’ effect penalises the choice of a university professional career vs. any other professional option, as well as the selection of a humanities and social science area training vs. a science one. To conclude, the analysis of the determining factors when selecting a professional career highlights that the humanities and social sciences have been penalised © 2012 Blackwell Publishing Ltd.

168 European Journal of Education, Part II in the last decade, with less relevance in the total number of PhD graduates. Besides, even if several factors determine the inclination for one training option or the other, it is the labour market expected wages that tend to move people away from humanities and social sciences. The professional status of the doctors of humanities and social sciences that work outside the university is especially worrying, as they are not satisfied with their job and with the low expected wages.This problem is very important if there is a wish to change doctoral education in Europe for it to be more firm-oriented because if people notice that in certain areas of knowledge (humanities and social sciences) the wages and labour conditions are penalised when moved from the university environment to the firm’s, then the number of doctors of this area will be dramatically reduced. Moreover, the great demand for doctors of science and the monetary incentive that the professional career outside the university represents may cause an ageing problem and loss of quality among university teachers in the long term. Juan Francisco Canal Domínguez. Department of Economy, University of Oviedo, Avenida del Cristo, s/n, 33071, Oviedo, Spain, [email protected] Manuel Antonio Muñiz Pérez. Department of Economy, University of Oviedo, Avenida del Cristo, s/n, 33071, Oviedo, Spain, [email protected] NOTES 1. This research was funded by the Project of the Ministry of Science and Innovation ECO2008-03468. We would like to thank INE (National Statistics Institute) for having provided us with the 2006 Survey on Human Resources in Science and Technology database. 2. There is no national directory of all those who hold a Doctorate. Hence, INE had to gather information from every university through the Consejo Superior de Universidades. To gather such information individually implied problems, as some of those universities do not have the lists, whilst others present heterogeneous lists in relation to their seniority, most of them being quite recent. 3. The following fields are included in the science area: natural sciences, engineering and technology, health sciences, agriculture sciences. Social sciences include economy, and business administration. The other doctors in social sciences are included in the humanities group, together with that of the humanities proper. 4. The National Statistical Institute provides data on trends in the number of new entrants by sex and field in Spain. 5. Data related to patents and inventions were not included, as they are nonexistent for the humanities and social sciences. REFERENCES Abbot, A. (2001) Chaos of Disciplines? (Chicago, the University of Chicago Press). Ben-David, J. (1992) Centers of Learning: Britain, France, Germany, United States (New Brunswick, Transaction Publishers). Brooks, R. & Heiland, D. (2007) Accountability, assessment and doctoral education: recommendations for moving forward, European Journal of Education, 42, pp. 351–362.

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

169

Buela-Casals, G. & Castro, A. (2008) Análisis de la evolución de los programas de doctorado con mención de calidad en las universidades españolas y pautas para su mejora, Revista de Investigación en Educación, 5, pp. 49–60. Clark, B. A. (1995) Places of Inquiry: research advanced education in modern universities (Berkeley, University of California Press). Crosier, D., Purser, L. & Smidt, H. (2007) Trends V: universities shaping the European Higher Education Area (Brussels, European University Association). Enders, J. (2004) Research training and careers in transition: a European perspective on the many faces of the Ph.D., Studies in Continuing Education, 26, pp. 419–430. Gibbons, M., Limoges, H., Nowotny, S., Schwartzman, P., Scott, P. & Trow, M. (1994) The New Production of Knowledge. The Dynamics of Science and Researching Contemporary Societies (London, Sage Publications). Ha¨ yrinen-Alestalo, M. & Peltola, U. (2006) The problem of a market-oriented university, Higher Education, 52, pp. 251–281 Jamieson, I. & Naidoo, R. (2007) University positioning and changing patterns of doctoral study: the case of the University of Bath, European Journal of Education, 42, pp. 363–373. Kehm, B. M. (2007) Quo vadis doctoral education? New European approaches in the context of global changes, European Journal of Education, 42, pp. 307–319. Mogue´ rou, P. (2005) Doctoral and postdoctoral education in science and engineering: Europe in the international competition, European Journal of Education, 40, pp. 367–392. Mora, J., Garci´a, J. & Garcia, A. (2000) Higher education and graduate employment in Spain, European Journal of Education, 2, pp. 229–237. Mora, J. G. & Vidal, J. (2000) Adequate policies and unintended effects in Spanish higher education, Tertiary Education and Management, 6, pp. 247–258. Naidoo, R. (2003) Repositioning higher education as a global commodity: opportunities and challenges for future sociology of education work, British Journal of Sociology of Education, 24, pp. 249–259. Naidoo, R. & Jamieson, I. (2005) Empowering participants or corroding learning? Towards a research agenda on the impact of student consumerism in higher education, Journal of Education Policy, 20, pp. 267–281. Noble, K. A. (1994) Changing Doctoral Degrees. An International Perspective (Bristol, Open University Press). Perotti, L. (2007) Institutional change in the Spanish higher education system, European Journal of Education, 42, pp. 411–423. Sanchez, L. (1996) Políticas de reforma universitaria en España: 1983–1993 (Madrid, Instituto Juan March). Schomburg, H. (2007) The professional success of higher education graduates, European Journal of Education, 42, pp. 35–57. Stephan, P. (1996) The economics of science, Journal of Economic Literature, 34, pp. 1199–1235

© 2012 Blackwell Publishing Ltd.

170 European Journal of Education, Part II APPENDIX Table AI. Definition of variables found in the different estimations Multinomial logit depending variable Professional career Variable taking value 1 if the person is a doctor of science and a university teacher, 2 if he is a doctor of science and has other professional activity, 3 if he is a doctor of humanities and a university teacher and 4 if he is a doctor of humanities and has other professional activity Wages equation depending variable Current earnings Earnings are specified in eight annual earning intervals logarithm Independent variables Worker characteristics: Age Age of worker Age2 Squared age of worker Male Dummy variable that takes value 1 if the worker is a man and 0 if the workers is a woman Married Dummy variable that takes value 1 if the worker is married and 0 otherwise Single Dummy variable that takes value 1 if the worker is single and 0 otherwise Other marital status Dummy variable that takes value 1 if the worker has a marital status which is not either married or single and 0 otherwise People under his Number of people who financially depend on the worker responsibility Training and research: PhD length Time from the beginning of doctoral studies until title is obtained Taking post-doc studies Dummy variable that takes value 1 if the worker is taking post-doc studies and 0 otherwise Published books Number of published books between 2003 and 2006 Published papers Number of published papers between 2003 and 2006 International mobility Dummy variable that takes value 1 if the worker has international mobility and 0 otherwise University Dummy variable that takes value 1 if the worker has become a doctor at this university and 0 otherwise Doctor of sciences Dummy variable that takes value 1 if the worker is a doctor of science and 0 if he is a doctor of humanities Job characteristics: Public sector Dummy variable that takes value 1 if firm’s activity belongs to the public sector and 0 if it belongs to the private sector Permanent contract Dummy variable that takes value 1 if the worker has a permanent contract and 0 if it is a temporal one Full time contract Dummy variable that takes value 1 if the worker has a full-time contract and 0 if it is part-time one Worked hours Number of hours worked during the week of reference PhD /job relation: high Dummy variable that takes value 1 if PhD /job relation is high and 0 otherwise PhD / job relation: Dummy variable that takes value 1 if PhD / job relation is normal, and 0 normal otherwise PhD / job relation: low Dummy variable that takes value 1 if PhD / job relation is low and 0 otherwise Minimum training level: Dummy variable that takes value 1 if minimum training level for a job is post post doctoral doctoral and 0 otherwise Minimum training level: Dummy variable that takes value 1 if minimum training level for a job is to be doctor a doctor and 0 otherwise Minimum training level: Dummy variable that takes value 1 if minimum training level for a job is to be graduate a graduate and 0 otherwise Minimum training level: Dummy variable that takes value 1 if minimum training level for a job is to be undergraduate a undergraduate and 0 otherwise Minimum training level: Dummy variable that takes value 1 if minimum training level for a job is to professional training have taken a professional training and 0 otherwise University teacher Dummy variable that takes value 1 if the worker is a university teacher and 0 if he has other professional activity Expected wages Expected annual wages logarithm

© 2012 Blackwell Publishing Ltd.

Juan Francisco Canal Domínguez & Manuel Antonio Muñiz Pérez

171

Table AII. Wage estimation

Constant Personal characteristics Age Age2 Male Married Single Other marital status Training and research Doctor of science Taking post-doc education Published books Published papers Job characteristics Public sector University teacher Permanent contract Full-time contract Worked hours No of observations: 12,193 c2 = 5,516.20 Prob > c2 = 0.00 The reference variable is Single.

© 2012 Blackwell Publishing Ltd.

Coefficient

Standard Error

1.484

14.97

0.056 -0.0001 0.107 0.072 0.089 0.024

12.22 -9.61 14.43 7.08 4.83 6.45

0.137 -0.043 0.004 0.001

15.56 -4.07 3.20 1.41

0.076 -0.042 0.245 0.231 -0.021

6.30 -5.30 24.03 10.66 -37.09