Links Between Parenting and Core Beliefs: Preliminary ... - Springer Link

C 2006), pp. 787–802 Cognitive Therapy and Research, Vol. 29, No. 6, December 2005 ( DOI: 10.1007/s10608-005-4291-6

Links Between Parenting and Core Beliefs: Preliminary Psychometric Validation of the Young Parenting Inventory Alex Sheffield,1,6 Glenn Waller,2 Francesca Emanuelli,3 James Murray,4 and Caroline Meyer5

This study describes the psychometric validation of the young parenting inventory (YPI), and tested specific hypotheses regarding the link between one’s experience of their parent’s behaviors and the development of schema-level core beliefs. The YPI is a measure of perceived parenting experiences, hypothesized to represent the origins of negative core beliefs. This preliminary validation consisted of analyses of factor structure, internal consistency, test–retest reliability, and construct validity. A large non-clinical student sample (N = 422) completed the YPI, and a subset also completed the Young Schema Questionnaire-Short form (YSQ-S). Factor analyses demonstrated that a shorter version of the questionnaire (YPI-R) could be developed to represent coherent and meaningful perceptions of each parent. The YPI-R consists of nine scales. Each scale had good test–retest reliability and adequate internal consistency. Significant associations between the YPI-R scales and negative core beliefs (as measured by the YSQ-S) indicated partial construct validity. At this preliminary stage, it can be concluded that the YPI-R has an acceptable level of psychometric utility. However, the hypothesized parenting-negative core belief links were not all substantiated. KEY WORDS: family functioning; parenting; core beliefs; validation.

Empirical research consistently reports a link between perception of one’s parenting and psychopathology (e.g., Murray, Waller, & Legg, 2000; Parker, 1983). A number of models have been suggested to explain this link. For example, attachment theory proposes that all infants have an innate motivation to seek proximity to their 1 Vincent

Square Clinic, London. George’s Hospital Medical School, University of London. 3 St. George’s Hospital Medical School, University of London. 4 University of Surrey. 5 University of Warwick. 6 Correspondence should be directed to Alex Sheffield: Vincent Square Clinic, Hopkinson House, Osbert St, London SW1; e-mail: [email protected]. 2 St.

787 C 2006 Springer Science+Business Media, Inc. 0147-5916/05/1200-0787/0

788

Sheffield, Waller, Emanuelli, Murray, and Meyer

primary caregivers, in order to achieve and maintain a sense of security (Bowlby, 1973; 1980). The emotional availability and responsiveness of the primary caregiver then determines the quality of the infant’s early attachment, which then influences the development of internal working models of the self and others. For example, a secure early attachment might result in a representation of oneself as worthy and competent, and of others as emotionally available and responsive, while an insecure attachment might lead to a representation of oneself as unworthy and incompetent and of others as rejecting and/or interfering. Attachment theory proposes that such representations then guide one’s interpretations, predictions and behavior in future relationships, and hence impact on adult psychopathology. An ongoing difficulty in exploring the parenting-psychopathology link is the reliability and validity of measures of parenting experiences. First, participants are required to report on something that occurred a significant time in the past, making recall bias likely. Second, the participant’s recall might be influenced by one’s more recent experience of their parents. Third, the participants recall may be influenced by filters currently in operation (e.g., schemas). The operations of current filters are likely to make it difficult for the person to reflect on their perceptions and experiences. For example, someone with a defectiveness belief (a belief that they are inherently flawed) might not be able to reflect on experiences of feeling criticized by their parents because, for them, it was just a “normal” experience. Finally, some participants might be using a variety of cognitive processes to defend against painful experiences (e.g., avoidance and idealization). Despite methodological limitations, empirical findings are consistent with the link between the quality of one’s early experiences (e.g., early attachment) and the development of later psychopathology, (e.g., Carlson, 1998; Warren, Huston, Egeland, & Sroufe, 1997). However, the link is not categorical. For example, not all insecurely attached infants will develop mental health problems, and not all securely attached infants will be protected from later psychological distress. Attachment theory might argue that this is because internal working models can be influenced by environmental changes throughout development (Bowlby, 1973). Therefore, although the parenting-psychopathology link is often supported, it continues to be deficient in two important respects. First, there is little clarity regarding the mechanism that links parenting to psychopathology. Second, there is a lack of specificity regarding the role of parenting in the development of different psychopathologies (e.g., Parker, 1983). It is suggested here that these deficits are due to poorly elaborated conceptualizations of the role of parenting in the development of psychopathology. Young (1999a; Young, Klosko, & Weishaar, 2003) has proposed a schemabased model to explain the parenting-psychopathology link. Young’s model draws on some of the concepts and research underpinning attachment theory but suggests that one potential mediator in the parenting-psychopathology link is the development of early maladaptive schemas or negative core beliefs, especially in Axis II and related pathology (e.g., impulsive problems). Negative core beliefs are unconditional, schema-level beliefs about the self, others or the world, (which overlap with the concept of internal working models). According to Young, such beliefs usually develop in childhood. One’s experience of being parented is therefore likely to be a key aetiological factor. Although negative core beliefs are appropriate to

Parenting and Core Beliefs

789

the environment in which they develop, they persist over time, failing to adapt to changing circumstances and leaving the person vulnerable to the development of psychological problems. For example, a child whose parents consistently criticized her might develop a defectiveness belief. That belief allows the child to make sense of her world at the time. However, if the belief persists into her adulthood, where those circumstances have changed, it is likely to impact on a variety of psychosocial factors. For example, the person might avoid close interpersonal relationships for fear of exposing their defectiveness and risking rejection. That avoidance may then lead to social isolation, leaving the person vulnerable to psychological distress. There is evidence of links between negative core beliefs and psychopathology in a number of areas, including the personality disorders (e.g., Ball & Cecero, 2001) and the eating disorders (e.g., Leung, Waller, & Thomas, 1999). The latter part of the model is therefore supported. However, in order to test the parenting-core belief link, an appropriate, model-driven measure of perceived parenting experiences is required. Research suggests that existing measures address levels of family functioning that are related to general psychopathology (e.g., Friedman et al., 1997). However, testing the parenting-core belief link requires a measure of parenting designed specifically to tap the parental origins of the development of negative core beliefs. Young has developed such a measure—the young parenting inventory (YPI; Young, 1999b). However, the YPI has yet to be validated. Therefore, the aim of this study was to perform a preliminary psychometric validation of the YPI in order to test the parenting-negative core belief link. According to Young’s model and the hypothesized structure of the YPI, it was anticipated that the YPI would consist of seventeen scales. It was hypothesized that each scale would be associated with the corresponding scale of Young’s measure of negative core beliefs, the Young Schema Questionnaire-short form (YSQ-S; Young, 1998). For example, the emotional deprivation scale of the YPI (which is thought to represent the origins of the emotional deprivation schema) should correlate positively with the emotional deprivation scale of the YSQ-S, the abandonment scale of the YPI (which is thought to represent the origins of the abandonment schema) should correlate positively with the abandonment scale of the YSQ-S, and so on. Preliminary validation of the YPI in this study will consist of analysis of factor structure, internal consistency, test–retest reliability and construct validity. Given that adverse early experiences and negative core beliefs are evident in non-clinical samples (although to a lesser extent than in clinical samples; Leung, Thomas, & Waller, 2000), this validation aimed to produce non-clinical norms, specifically in preparation for comparison with later clinical samples.

METHOD Participants The sample consisted of 422 undergraduate and postgraduate University students. Of these, 68 (16.2%) were male, 353 (83.6%) were female, and one failed to report his/her gender. The mean age of the sample was 24.5 years (SD = 7.9, range = 18–61). All participants completed the YPI. For the purpose of determining

790


construct validity, some participants were randomly selected to complete a further questionnaire, measuring negative core beliefs (the Young Schema QuestionnaireShort form). Of the total sample, this produced 160 participants who completed both the YPI and the YSQ-S. Therefore, the number of participants varies between analyses. Measures Young Parenting Inventory (YPI; Young, 1999b). The YPI was designed on the basis of clinical experience, and is intended to identify the potential origins of seventeen negative core beliefs. It is a self-report questionnaire consisting of 72 statements that individuals might use to describe their parents (e.g., “criticized me a lot”). Each statement reflects parental behaviors that are related to one of the following seventeen negative core beliefs: emotional deprivation—the belief that one’s emotional needs will not be met by others; abandonment—close relationships will always end; mistrust/abuse—one will be taken advantage of or abused by others; vulnerability to harm—one has no control over threats of illness or disaster; defectiveness/shame—one is inherently flawed; failure to achieve— one is incapable of succeeding; subjugation—one’s opinions, desires and feelings are not as important as others’; self-sacrifice—one must always put others first; dependence/incompetence—one is not independent, and cannot cope with everyday tasks; unrelenting standards—one must reach impossibly high standards; entitlement—one is entitled to anything without consideration for others; insufficient self-control/self discipline—one is unable to control impulses or feelings; enmeshment–lack of identity, as a result of emotional overinvolvement with others; negativity/pessimism—nothing will ever turn out as one would like; emotional inhibition—one should hide emotions; punitiveness—one will always be punished; and approval-seeking—the ongoing desire to gain approval from others. Each of the 72 items is rated on two six-point Likert scales, reflecting how well each statement describes the participant’s mother and father. With the exception of the emotionally depriving scale, which is reverse scored, 1: “completely untrue” and 6: “describes him/her perfectly.” Higher scores indicate a perception that the parent behaved in ways that were more likely to generate the related core beliefs. Young Schema Questionnaire—Short form (YSQ-S; Young, 1998). The YSQ-S is a 75-item self-report questionnaire, assessing the level of 15 negative core beliefs. Those core beliefs include all those outlined in the YPI (above), but do not include negativity/pessimism, punitiveness and approval-seeking.7 Participants rate their beliefs on six-point Likert scales (ranging from 1: “completely untrue of me” to 6: “describes me perfectly”). In all cases higher scores indicate more pathological core beliefs. The YSQ-S has been validated clinically and psychometrically in a range of populations (e.g., Stopa, Thorne, Waters, & Preston, 2001; Waller, Meyer, & Ohanian, 2001; Welburn, Coristine, Dagg, Ponefract, & Jordan, 2002). 7 One

of the schemas measured by the YSQ-S scale is social isolation/alienation. The origins of this schema are not measured in the YPI, because Young hypothesises that the origins are usually in the peer group.


791

Procedure The relevant ethics boards approved the study. A variety of classes were randomly selected across a broad range of university courses. The participants were approached during lectures, and were asked to complete the measures and return them in prepaid envelope. With the aim of producing a test–retest sample of approximately 40 test–retest participants, two classes were randomly selected8 and those participants were asked to complete a second copy of the YPI four weeks later, to determine test–retest reliability. This resulted in 31 test–retest participants. Each participant completed the appropriate consent form. Data Analysis Factor analyses were used to determine the factor structures of the YPI for each of the two parents (direct oblimin method, to allow for likely intercorrelation of scales). There was a very substantial overlap of items loadings across the two factor structures, allowing the generation of a single overlapping factor structure (see results). Cronbach’s alpha was used to assess the internal consistency of items in the resulting scales. As some of the resulting scales were not sufficiently normally distributed to allow for parametric analyses, the remaining data analyses were conducted using non-parametric methods. Test–retest reliability was determined through two-tailed Wilcoxon tests and Spearman’s rho correlations. Construct validity was assessed using Spearman’s rho correlations to determine the relationships between the YPI-R and the YSQ-S. RESULTS Factor Structure and Internal Reliability For each factor analysis, factors were included if they had an eigenvalue of greater than 1.00, and items were included if they had a loading of 0.40 or above (Comrey & Lee, 1992). A small number of further items were excluded if they had poor face validity. There were 11 maternal factors, with eigenvalues ranging from 1.04 to 19.43 (percentage of variance explained = 1.77–26.98). There were 14 paternal factors, with eigenvalues ranging from 1.16 to 19.35 (percentage of variance explained = 1.54–26.87). Because there was a substantial overlap between the two factor structures, an overlapping scale was generated, using only those items and factors that were common to both parents. In order for a scale to be included in the revised YPI (YPI-R), there was a requirement for the Cronbach’s alpha of the items to be greater than 0.65 for both parents (Nunnally, 1978). The scale also had to show strong test–retest reliability for both parents (non-significant Wilcoxon test and a significant correlation between scores across time). Table I shows the final combined factor structure and Cronbach’s alphas for each scale. 8 Formal

randomization was not used in the selection process.

Belittling parenting—This scale reflects behavior in the parent that belittles the child, leaving the child feeling defective. A high score indicates a greater intensity of belittling behavior by the parent. A low score indicates that the parent was not belittling in his/her behavior.

Overprotective parenting—This scale reflects a pattern of parenting that overprotects the child. A high score indicates that the parent is overprotective, worrying excessively and failing to foster the child’s independence. A low score indicates that the parent is not overprotective.

Emotionally depriving parenting—This scale reflects a pattern of parenting that deprives the child of emotional nurturing. The item scores are reversed. Thus, a high score reflects a failure by the parent to emotionally nurture the child. A low score reflects appropriate emotional nurturing by the parent.

YPI-R scale and description of scale Loved me, treated me as someone special. Spent time with and paid attention to me. Gave me helpful guidance and direction. Listened to me, understood me, shared feelings with me.

1a

3a 4a

Overprotected me. Made me feel I couldn’t rely on my decisions or judgment. Did too many things for me instead of letting me do things on my own. Treated me as if I were younger than I really was.

17 18

19 20

Made me feel unloved or rejected.

Treated me as if there was something wrong with me.

22

23

15

Worried excessively that I would get hurt. Worried excessively that I would get sick

14

2a

Item

No.

.792

Paternal

.747 .700ed,pf

Paternal Maternal

.725

.675ct .709 Paternal Maternal

Maternal

.523bl,cn .616ct

−.644ei .808 .806 .799 Paternal Maternal Paternal Maternal

Paternal Maternal

−.685 −.620ei

Paternal Maternal

.789 .752 .793 .528cn,pf

−.739 −.612bl

Paternal Maternal

Paternal Maternal Paternal Maternal

−.611bl,pf −.679bl −.645bl,pf

Loading

Maternal Paternal Maternal

Maternal/paternal statistic

Table I. Overlapping Factor Structure of the YPI and Test–Retest Analysis of the Revised YPI Scales

.91

.84 .84

.91 .92

Cronbach’s alpha

Pessimistic/fearful parent—This scale reflects anxious, fearful traits in the parent. A high score indicates an angst-ridden parent, with a pessimistic outlook on life. A low score indicates the absence of these traits.

Perfectionist parenting—This scale reflects the parents’ expectations for themselves as well as for their child. A high score indicates that the parent expected perfection of themselves and of their child. A low score indicates that the parent had more relaxed standards for themselves and their child.

Treated me as if my opinions or desires didn’t count.

29

Was a fearful or phobic person

Worried a lot about the family’s financial problems. Had a pessimistic outlook; often expected the worst outcome. Focused on the negative aspects of life or things going wrong.

16

56 58 59

38 Was a perfectionist in many areas; things had to be ‘just so’

Expected me to be a failure in life.

28

39

Didn’t really want me to succeed.

27

Had very high expectations for him/herself. Expected me to do my best at all times

Treated me as if I was stupid or untalented.

26

37

Made me feel ashamed of myself in important respects.

24

.714ct .596 .725 .825 .673ct .827 .748 .885 .747ct .791ct .764 .622 .678 .566 .648 .787 .535

.507 .403 .458 .824 .834 .831 .830

Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal Maternal

Paternal Maternal Paternal Maternal Paternal Maternal Paternal

.73

.77

.67 .69

I felt that I didn’t have my own sense of direction while I was growing up because he/she was such a strong person. Was warm and physically affectionate

54

Punitive parenting—This scale reflects a pattern of parenting that is punitive of the child’s mistakes. A high score indicates a greater level of punitive behavior. A low score indicates low levels of punitive behavior.

67

66

Would call me names (like “stupid” or “idiot”) when I made mistakes.

Would become angry or harshly critical when I did something wrong. Would punish me when I did something wrong.

Was private; rarely discussed his/her feelings.

64

65

Was uncomfortable expressing affection or vulnerability.

61

5a

I felt that I didn’t have enough individuality or sense of self separate from him/her.

53

Emotionally inhibited parenting—This scale reflects the parent’s ability to share their feelings with their child. A high score indicates that they do not share their feelings with their child. A low score indicates that they are able to share their feelings with their child.

Controlled my life so that I had little freedom of choice.

31

Controlling parenting—This scale reflects a pattern of parenting that controls or inhibits the child’s independence. A high score indicates that the parent is highly controlling, limiting the development of the child’s sense of self. A low score indicates that the parent is not controlling, but rather allows the child to become independent and make their own decisions.

Item

No.

YPI-R scale and description of scale

Table I. Continued

.792 −.765 .564bl,cn −.761

Paternal Maternal Paternal

.807 .578bl,cn −.778bl

Paternal Maternal Paternal Maternal

.739 .620

.693 −.772 .500

Paternal Maternal Paternal

Paternal Maternal

−.759

Paternal Maternal

.74 .79

.71 .80

.70

.716pu,bl

−.573bl,ed −.541ed .744

.78

−.704bl,cn

Maternal

Maternal Paternal Maternal

Cronbach’s alpha

Loading

Maternal/paternal statistic

Was concerned with social status and appearance. Placed strong emphasis on success and competition. Was concerned with how my behavior would reflect on him/her in the eyes of others. Seemed to love me more or pay more attention to me when I excelled.

69 70 71

72

.845 .827 .723pt .703 .779 .747 .572 .577

Maternal Paternal Maternal Paternal Maternal Paternal Maternal Paternal

.77

.79

Note. Only the highest loading is recorded for each item. Where a superscript is inserted after a loading, it indicates that this item also loads (>.4,) on another factor/s: ed: emotionally depriving parenting; bl: belittling parenting; ct: controlling parenting; pu: punitive parenting; cn: conditional/narcissistic parenting; pf: pessimistic/fearful parenting; ei: emotionally inhibited parenting; pt: perfectionist parenting. a These items were reverse scored.

Conditional/narcissistic parenting—This scale reflects behavior by the parent that implies that positive regard for the child is conditional on the child’s success. A high score indicates a high level of conditional acceptance by the parent. A low score indicates that positive regard for the child is unconditional.

796


To summarize, the YPI-R consisted of nine factors that were common to both parents: emotionally depriving parenting; overprotective parenting; belittling parenting; perfectionist parenting; pessimistic/fearful parenting; controlling parenting; emotionally inhibited parenting; punitive parenting; and conditional/narcissistic parenting. These factors had good internal consistency. In order to score these scales, the mean score is calculated for those items that are in the scale. Test–Retest Reliability Table II shows the findings of the test–rest reliability analysis. All nine scales have good test–retest reliability.

Table II. Test–Retest Reliability (N = 31) for YPI-R scales Time 1 YPI factors

Time 2

Wilcoxon test

Spearman’s correlation

M

SD

M

SD

Z

P

ρ

P

1.55 2.19

0.60 0.84

1.64 2.19

0.72 0.80

0.02 0.52

ns ns

0.81 0.77

.001 .001

1.99 1.88

1.08 1.07

1.99 1.86

1.03 1.04

0.47 0.52

ns ns

0.85 0.75

.001 .001

1.08 1.17

0.21 0.39

1.08 1.09

0.17 0.23

0.37 1.19

ns ns

0.61 0.53

.001 .003

3.73 3.93

1.22 1.18

4.01 4.07

1.13 1.20

1.62 0.49

ns ns

0.65 0.67

.001 .001

Emotionally depriving Mothers Fathers Overprotective Mothers Fathers Belittling Mothers Fathers Perfectionist Mothers Fathers Pessimistic/fearful Mothers Fathers Controlling Mothers Fathers

1.92 1.81

0.76 0.70

1.85 1.82

0.75 0.87

0.27 0.18

ns ns

0.53 0.69

.003 .001

1.37 1.36

0.50 0.55

1.55 1.48

0.83 0.86

0.86 0.86

ns ns

0.65 0.69

.001 .001

Emotionally inhibited Mothers Fathers

2.64 3.64

1.02 1.41

2.83 3.65

1.01 1.17

1.46 0.11

ns ns

0.81 0.73

.001 .001

Punitive Mothers Fathers

2.10 2.27

1.00 1.24

1.96 2.13

1.06 1.30

0.85 1.15

ns ns

0.62 0.78

.001 .001

Conditional/narcissistic Mothers Fathers

2.39 2.50

1.28 1.41

2.27 2.60

0.24 1.40

0.37 1.06

ns ns

0.84 0.80

.001 .001


797 Table III. Mean Scores on the YPI-R YPI-R scales

Full sample M (SD)

Emotionally depriving mothers Emotionally depriving fathers Overprotective mothers Overprotective fathers Belittling mothers Belittling fathers Perfectionist mothers Perfectionist fathers Pessimistic/fearful mothers Pessimistic/fearful fathers Controlling mothers Controlling fathers Emotionally inhibited mothers Emotionally inhibited fathers Punitive mothers Punitive fathers Conditional/narcissistic mothers Conditional/narcissistic fathers

1.91 (1.04) 2.39 (1.25) 2.09 (1.00) 1.88 (0.93) 1.28 (0.67) 1.35 (0.77) 3.47 (1.16) 3.63 (1.24) 2.10 (1.12) 1.93 (0.97) 1.61 (0.90) 1.51 (0.83) 2.69 (1.20) 3.35 (1.31) 2.13 (1.07) 2.25 (1.22) 2.56 (1.17) 2.56 (1.19)

Note. Full sample N = 392.

Construct Validity As the YPI and YPI-R are intended to represent potential origins of negative core beliefs, Spearman’s rho correlations were calculated to test the association of YPI-R factors with negative core beliefs (measured by the YSQ-S). Tables III and IV respectively, give the sample’s mean scores on each and YPI-R and YSQ-S scale. Table V shows the associations (Spearmans rho) between the YPI-R scales and the YSQ-S scales. Due to the large number of correlations, a Bonferonni correction was used (P < .00002) to reduce the risk of Type 1 errors. Both maternal and

Table IV. Mean Scores on the YSQ-S YSQ-S Scale

Full sample M (SD)

Emotional deprivation Abandonment Mistrust/abuse Social isolation Defectiveness shame Failure Dependence/incompetence Vulnerability to harm Enmeshment Subjugation Self-sacrifice Emotional inhibition Unrelenting standards Entitlement Insufficient self control/self discipline

1.93 (1.09) 2.20 (1.19) 2.21 (1.04) 2.05 (1.11) 1.63 (0.93) 1.90 (1.06) 1.68 (0.75) 1.81 (0.87) 1.53 (0.76) 1.86 (0.88) 3.08 (1.17) 2.13 (1.16) 3.80 (1.17) 2.37 (0.87) 2.38 (1.06)

Note. Full sample N = 242.

.499∗ .454∗

Belittling Mothers Fathers

.255∗ .208

.321∗ .257∗

.277∗ .353∗

.235 .188

Controlling Mothers Fathers

Emotionally inhibited Mothers Fathers

Punitive Mothers Fathers

Conditional/ narcissistic Mothers Fathers

.184 .163

.235 .316∗

.156 .180

.245∗ .171

.277∗ .222

−.005 .083

.284∗ .288∗

.133 .109

.199 .145

Note. ∗ p < .00002; N = 229–242.

.459∗ .218

Pessimistic/ fearful Mothers Fathers

−.071 −.096

.140 .127

Overprotective Mothers Fathers

Perfectionist Mothers Fathers

.527∗ .377∗

.204 .279∗

.282∗ .322∗

.227 .247∗

.262∗ .263∗

.246∗ .182

.081 .157

.315∗ .285∗

.292∗ .267∗

.154 .100

.166 .133

.230 .238

.272∗ .231

.247∗ .198

.300∗ .170

.034 .066

.289∗ .296∗

.219 .166

.277∗ .186

.145 .118

.221 .192

.247∗ .155

.256∗ .219

.249∗ .162

.094 .085

.252∗ .281∗

.173 .176

.135 .120

.177 .223

.205 .233

.167 .202

.334∗ .325∗

.223 .206

.060 .088

.275∗ .236

.240 .197

.188 .175

.139 .130

.236 .307∗

.089 .145

.231 .216

.237 .151

.026 −.002

.194 .193

.208 .148

.148 .160

.175 .139

.187 .256∗

.064 .082

.205 .190

.275∗ .218

.055 .005

.247∗ .231

.300∗ .251∗

.102 .090

.106 .174

.087 .121

.070 .126

.309∗ .223

.237 .235

.004 .004

.046 .179

.401∗ .266∗

−.064 .121

.269∗ .216

.295∗ .297∗

.291∗ .151

.347∗ .275∗

.330∗ .158

.170 .073

.307∗ .289∗

.332∗ .228

.256∗ .211

.180 .147

.165 .176

.180 .087

.196 .189

.109 .077

.186 .225

.198 .082

.227 .240

.156 .015

.206 .212

.206 .145

.258∗ .172

.271∗ .235

.182 .085

.120 .123

.298∗ .228

.201 .146

.267∗ .164

.275∗ .331∗

.162 .207

.317∗ .209

.188 .202

.193 .175

.222 .223

.284∗ .241

.157 .167

.188 .100

.364∗ .383∗

.247∗ .263∗

.180 .140

.207 .183

.234 .139

.133 .179

.178 .158

.215 .191

.130 .117

Emotional Abandon- Mistrust/ Social Defectiveness/ Dependence/ Vulnerability Enmesh- SubjugSelf Emotional Unrelenting Entitledeprivation ment abuse isolation shame Failure incompetence to harm ment ation sacrifice inhibition standards ment

Emotionally depriving Mothers Fathers

YPI-R scales

Correlation with YSQ scales

Table V. Association (Spearman’s rho) Between the YPI-R Scales and the YSQ-S Scales1,2

.290∗ .271∗

.272∗ .234

.159 .152

.293∗ .220

.227 .115

.073 .045

.195 .178

.214 .213

.141 .149

Insufficient self control/self discipline


799

paternal YPI-R scales were commonly associated with a number of YSQ-S scales. Many of the associations were clinically meaningful (for example, belittling mothers correlated with a failure belief; punitive mothers and fathers correlated with a mistrust/abuse belief). All YPI-R scales correlated significantly with some YSQ-S scales with the exception of Perfectionist mothers and fathers and pessimistic/fearful fathers. Similarly, all YSQ-S scales were significantly correlated with some YPI-R scales, with the exception of Self-sacrifice. The YPI-R scales that were associated with greatest number of negative core beliefs (i.e., consistent with having the broadest effect) included belittling mothers and fathers, pessimistic/fearful mothers, controlling mothers and punitive fathers. Similarly, some negative core beliefs were associated with many negative parenting styles (e.g., emotional deprivation, mistrust/abuse, and subjugation). To summarize, given the significant correlations between YPI-R and YSQ-S scales, the results are consistent with the notion that the YPI-R measures potential origins of negative core beliefs. However, they YPI-R scales do not map neatly or specifically to the beliefs that were proposed for purposes of construct validity. For example, it is not the case that the emotional deprivation scale of the YPI-R correlates only with the emotional deprivation scale of the YSQ-S, and so on.

DISCUSSION This study has conducted a preliminary validation of the YPI—a measure designed to explain those parental behaviors that account for the development of negative core beliefs (Young, 1999b). It then tested specific hypotheses regarding the link between parenting and the development of negative core beliefs. In the validation phase, factor analyses demonstrated that a shorter version of the questionnaire (YPI-R) could be developed and applied to perceptions of each parent. The YPI-R consists of nine scales, with a total of 37 items. The scales were: emotionally depriving parenting; overprotective parenting; belittling parenting; perfectionist parenting; pessimistic/fearful parenting; controlling parenting; emotionally inhibited parenting; punitive parenting; and conditional/narcissistic parenting. Each of these scales had good test–retest reliability and adequate internal consistency. Significant and clinically meaningful correlations between the YPI-R and YSQ-S scales are consistent with the general hypothesis that the YPI-R measures potential origins of negative core beliefs. It can therefore be concluded that the YPI-R has an acceptable level of psychometric validity, and that it can potentially be used to explain part of the variance in negative core beliefs. In the hypothesis-testing phase, YPI-R scales were related to all 15 negative core beliefs (as measured by the YSQ-S, Young, 1998) with the exception of the self-sacrifice scale. This suggests that either the YPI-R scales do not measure the origins of a self-sacrifice belief, or that the origins of this belief do not lie in one’s experience of being parented. Similarly, the absence of significant correlations with Perfectionist mothers and fathers and the Emotionally Inhibited fathers scales of the YPI-R suggests that these particular experiences of being parented do not represent potential origins of the negative core beliefs measured by the YSQ-S.

800


Both maternal and paternal YPI-R scales were significantly correlated with YSQ-S scales. Although there were a number of clinically meaningful correlations, the correlations were not in the linear pattern predicted. For example, it was not the case that high scores on a particular YPI-R scale (e.g., emotionally depriving mothers/fathers) correlated only with its corresponding YSQ-S scale (e.g., emotional deprivation). Although this might be partly because some YPI-R scales differed from the original YPI scales, it is also consistent with the notion that perceptions of parenting can place the individual at a greater risk of developing a variety of negative core beliefs, rather than a single negative core belief. Also, some particular perceptions of parenting (including belittling mothers, controlling mothers, controlling fathers and punitive fathers) correlated with a greater number of YSQ-S scales. This is consistent with them having a broader effect on the development of negative core beliefs than other perceptions of parenting. Similarly, some negative core beliefs (including emotional deprivation, failure, and subjugation) correlated with a greater number of parenting scales. In other words, the origins of some negative core beliefs are present in a variety of perceptions of parenting. To summarize, although there are some variations, the results appear to be more consistent with parenting having a general effect on negative core beliefs, as well as some specific linkage. This is consistent with the general pattern of findings from other studies. For example, studies using the parental bonding instrument (Parker, Tupling, & Brown, 1979) as a measure of parenting experiences show a broad pattern of lower care and greater overprotection in clinical groups in general (e.g., Murray et al., 2000). The value of the YPI-R would then lie in its ability to explain the specific core beliefs that are prominent in individual disorders, as well as those that are common to disorders in general (e.g., defectiveness/shame, which seems to underpin many cases). Such an understanding would allow the clinician to understand the origins of the core beliefs and yet challenge the need for such beliefs to be maintained in the current environment. It is important to note that some YSQ-S scales might measure maladaptive coping strategies rather than negative core beliefs. For example, a person may develop a subjugation belief because they believe that if they subjugate themselves then they can avoid others abandoning them, and hence avoid the associated negative cognitions and emotions associated with feeling abandoned. If this is the case, it is likely that the development of maladaptive coping strategies may also lie in one’s experience of being parented. However, it is impossible to disentangle the measurement of primary negative core beliefs and the development of maladaptive coping strategies in a study of this kind. While this study has provided preliminary validation of the YPI-R, and evidence of the link between parenting and negative core beliefs, it requires replication, extension and further validation. It is likely that a non-clinical sample will have lower levels of negative core beliefs and lower levels of adverse parenting experiences, and hence relationships between key variables are less likely to emerge. It is therefore necessary to test the utility of the YPI-R among clinical groups where family function and core beliefs have been hypothesized to be relevant to the etiology and maintenance of the disorder (e.g., Parker, 1983; Young et al., 2003). Although the YPI-R scales reflect relatively distinct constructs, they are likely to reflect


801

parental behaviors that co-occur. Therefore, multivariate methods (e.g., multiple regression) will be necessary to develop a more accurate perspective on the role of parental behaviors. The YPI-R might also be tested more specifically for its association with maladaptive coping strategies, used to avoid the activation of negative cognitions and emotions (e.g., Young et al., 2003). The YPI-R will need to be compared with other measures of family function, to determine whether its specificity means that it has greater predictive validity than other measures. Given the role of core beliefs in Axis II pathology (Young, 1999a; Young et al., 2003), it can be hypothesized that the YPI-R will have particular value in explaining disorders with an element of personality pathology or comorbidity with personality disorders. Prospective designs will also assist in determining the clinical utility of these findings. If that clinical utility can be established, then the YPI-R will be a useful tool in explaining part of the development of schema-level cognitions, including the development of the negative core beliefs that predispose the individual to psychopathology. However, it will still be necessary to develop an understanding of the role of non-parental environmental factors and temperamental factors in the development of such belief systems across childhood and adolescence (e.g., Young et al., 2003).

REFERENCES Ball, S. A., & Cecero, J. J. (2001). Addicted patients with personality disorders: Traits, schemas, and presenting problems. Journal of Personality Disorders, 15, 72–83. Bowlby, J. (1973). Attachment and loss: Vol. 2. Separation. New York: Basic Books. Bowlby, J. (1980). Attachment and loss: Vol. 3. Loss. New Scale. Philadelphia: Lippincott. Carlson, E. A. (1998). A prospective longitudinal study of disorganized/disoriented attachment. Child Development, 69, 1107–1128. Comrey, A. L., & Lee, H. B. (1992). A first course in factor analysis (2nd Ed.). Hillsdale, NJ: Erlbaum. Friedman, M. S., McDermut, W. H., Solomon, D. A., Ryan, C. E., Keitner, G. I., & Miller, I. W. (1997). Family functioning and mental illness: a comparison of psychiatric and non-clinical families. Family Process, 36, 357–367. Leung, N., Thomas, G. V., & Waller, G. (2000). The relationship between parental bonding and core beliefs in anorexic and bulimic women. British Journal of Clinical Psychology, 39, 203–213. Leung, N., Waller, G., & Thomas, G. V. (1999). Core beliefs in anorexic and bulimic women. Journal of Nervous and Mental Disease, 187, 736–741. Murray, C., Waller, G., & Legg, C. (2000). Family dysfunction and bulimic psychopathology: The mediating role of shame. International Journal of Eating Disorders, 25, 319–326. Nunnally, J. C. (1978). Psychometric theory. New York: McGraw Hill. Parker, G. (1983). Parental overprotection: A risk factor in psychosocial development. New York: Grune & Stratton. Parker, G., Tupling, H., & Brown, L. B. (1979). A parental bonding instrument. British Journal of Medical Psychology, 52, 1–10. Stopa, L., Thorne, P., Waters, A., & Preston, J. (2001). Are the short and long forms of the Young Schema Questionnaire comparable and how well does each version predict psychopathology scores? Journal of Cognitive Psychotherapy, 15, 253–272. Waller, G., Meyer, C., & Ohanian, V. (2001). Psychometric properties of the long and short versions of the Young Schema Questionnaire: Core beliefs among bulimic and comparison women. Cognitive Therapy and Research, 25, 137–147. Warren, S. L., Huston, L., Egeland, B., & Sroufe, L. A. (1997). Child and adolescent anxiety disorders and early attachment. Journal of the American Academy of Child and Adolescent Psychiatry, 36, 637–644. Welburn, K., Coristine, M., Dagg, P., Ponefract, A., & Jordan, S. (2002). The Schema QuestionnaireShort Form: Factor analysis and relationship between schemas and symptoms. Cognitive Therapy and Research, 26, 519–530.

802


Young, J. E. (1998). Young Schema Questionnaire–short form (YSQ-S) (On-line). New York: Cognitive Therapy Centre. (Available: http://www.schematherapy.com). Young, J. E. (1999a). Cognitive therapy for personality disorders: A schema focused approach (3rd ed.). Sarasota, FL: Professional Resource Exchange. Young, J. E. (1999b). Young Parenting Inventory (YPI) (On-line). New York: Cognitive Therapy Centre. (Available: http://www.schematherapy.com). Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner’s guide. New York: Guilford.