Child Abuse & Neglect 38 (2014) 735–746

Contents lists available at ScienceDirect

Child Abuse & Neglect

Emotional abuse as a predictor of early maladaptive schemas in adolescents: Contributions to the development of depressive and social anxiety symptoms夽 E. Calvete ∗ University of Deusto, Spain

a r t i c l e

i n f o

Article history: Received 19 June 2013 Received in revised form 14 October 2013 Accepted 16 October 2013 Available online 16 November 2013

Keywords: Parental emotional abuse Bullying victimization Temperament Early maladaptive schemas Depression Social anxiety Adolescents

a b s t r a c t The schema therapy model posits that maltreatment generates early maladaptive schemas (EMSs) that lead to the development of emotional disorders throughout the life span. The model also stipulates that temperament moderates the influence of maltreatment on EMSs. This study examines (a) whether emotional abuse perpetrated by parents and peers, both alone and interactively with temperament, predicts the worsening of EMSs; and (b) whether EMSs in turn predict an increase in depressive and social anxiety symptoms in adolescents. A total of 1,052 adolescents (Mage = 13.43; SD = 1.29) were assessed at three time points, each of which was separated by 6 months. The subjects completed measures of emotional abuse by parents and peers, neuroticism, extraversion, EMSs, depressive symptoms, and social anxiety. The findings indicate that emotional bullying victimization and neuroticism predict a worsening of all schema domains over time. Contrary to expectations, there was no significant interaction between temperament dimensions and emotional abuse. The results confirmed the mediational hypothesis that changes in EMSs mediated the predictive association between bullying victimization and emotional symptoms. This study provides partial support for the schema therapy model by demonstrating the role of emotional abuse and temperament in the genesis of EMSs. © 2013 Elsevier Ltd. All rights reserved.

Introduction Childhood maltreatment is a strong predictor of psychological disorders, such as depression and anxiety, during adolescence and throughout the life span (e.g., Alloy, Abramson, Smith, Gibb, & Neeren, 2006; Cicchetti & Valentino, 2006; Hankin, 2005; Harkness, Bruce, & Lumley, 2006; Kim & Cicchetti, 2010; Simon et al., 2009). Specifically, emotional abuse seems to be particularly relevant for the development of depression (Gibb, Butler, & Beck, 2003; Gibb, Chelminski, & Zimmerman, 2007) and social anxiety (Bruce, Heimberg, Blanco, Schneier, & Liebowitz, 2012; Simon et al., 2009). In addition to maltreatment perpetrated by parents, abuse perpetrated by peers has also been identified as a risk factor for the development of depression and social anxiety (e.g., Cole et al., 2013; Reijntjes, Kamphuis, Prinzie, & Telch, 2010). Cognitive theories posit that the impact of childhood maltreatment on subsequent psychopathology may be mediated by cognitive vulnerabilities, which include negative inference styles and dysfunctional schemas (Gibb, Abramson, & Alloy, 2004; Hankin, 2005). As support for this assumption, diverse models hold that maltreatment experiences contribute to the development of cognitive vulnerabilities (Ingram, 2003). The influence of maltreatment on cognitive vulnerabilities would

夽 This research was supported by the Ministerio de Ciencia e Innovación (Spanish Government, Ref. PSI2010-15714). ∗ Corresponding author address: Department of Personality, Psychological Assessment and Treatment, University of Deusto, Apdo. 1, 48080-Bilbao, Spain. 0145-2134/$ – see front matter © 2013 Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.chiabu.2013.10.014

736

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

be particularly strong when early experiences interact with specific temperament dimensions, an aspect that has received scant attention so far. The present study expands these ideas to the concept of early maladaptive schemas (EMSs), a central construct of schema therapy (Young, Klosko, & Weishaar, 2003), and explores the role of emotional abuse, both alone and in interaction with temperament, in the development of EMSs and depression and social anxiety symptoms in adolescents. The schema therapy model Schema therapy is an integrative model of therapy developed by Young and colleagues (Young, 1999; Young et al., 2003) that has been applied to several psychological problems, including depression and anxiety (e.g., Hinrichsen, Waller, & Emanuelli, 2004; Rijkeboer, van den Bergh, & van den Bout, 2005). EMSs constitute the key construct within this model. EMSs are defined as broad, dysfunctional, and pervasive patterns consisting of memories, emotions, cognitions, and bodily sensations about oneself and relationships with others (Young et al., 2003). EMSs are hypothesized to originate early in childhood and to be elaborated on throughout one’s lifetime under the influence of experiences and temperament. The schema theory proposes the existence of 18 schemas grouped into five broad categories of unmet emotional needs or schema domains. Recent studies conducted with adolescents indicate that three of these schema domains (disconnection/rejection, impaired autonomy and performance, and other-directedness) predict depressive and social anxiety symptoms (Calvete, Orue, & Hankin, 2013a, 2013b). The disconnection/rejection domain includes schemas such as abuse and defectiveness, which involve the expectation that one’s needs for acceptance and respect will not be predictably fulfilled. The domain of impaired autonomy consists of expectations about oneself and the environment that interfere with one’s perceived capacity to function independently or perform successfully. It includes the failure schema, which describes the belief that one has failed, will inevitably fail, or is fundamentally inadequate relative to one’s peers in areas of achievement. Finally, the schemas within the domain of other-directedness consist of an extreme focus on others’ desires at the expense of one’s own needs (e.g., the need for acceptance and subjugation). The disconnection/rejection and the impaired autonomy and performance domains predict the increase of depressive symptoms (Calvete et al., 2013a), whereas the disconnection/rejection and other-focused domains are predictive of increases in social anxiety (Calvete et al., 2013b). The role of maltreatment in the development of EMSs Young et al. (2003) theorized that EMSs originate as the result of the interaction between early experiences and the child’s temperament. In particular, those experiences that prevent the child from satisfying emotional needs would play a central role in the development of EMSs. According to their model, a relevant experience that can contribute to the development of EMSs is victimization. The child who is harmed or victimized within the family can develop schemas such as abuse, defectiveness, or vulnerability to harm. Several cross-sectional studies have found significant associations between maltreatment perpetrated by parents and EMSs (Calvete & Orue, 2013; Carr & Francis, 2010; McCarthy & Lumley, 2012; Muris, 2006; Thimm, 2010a; Wright, Crawford, & Del Castillo, 2009). Overall these studies indicate that experiences of emotional maltreatment and neglect are mainly associated with schemas within the disconnection/rejection and impaired autonomy domains. Nevertheless, the vast majority of these studies were conducted with adult samples (for exceptions see Calvete & Orue, 2013; Muris, 2006), and the participants were asked to retrospectively report early experiences of maltreatment and dysfunctional parenting. This adult focus and retrospective reporting limits the validity of the results because current emotional states and schemas could have biased memories of past maltreatment events. In contrast with the lack of prospective evidence for the link between parental maltreatment and EMSs, a few longitudinal studies have shown how parental emotional abuse predicts other cognitive vulnerabilities, such as inferential styles (Gibb & Abela, 2008; Padilla & Calvete, in press) and ruminative responses (Padilla & Calvete, in press). Although EMSs are hypothesized to originate early in childhood, Young’s model posits that they are further elaborated on throughout one’s lifetime under the influence of new experiences (Young et al., 2003). Thus, in addition to family, other influences, such as peers and school, become increasingly important as the child matures and may contribute to the construction of schemas. This continuous process of construction of schemas is supported by the scarce studies on EMSs in childhood, which suggest that schemas are present in children (Stallard & Rayner, 2005) but are still moderately stable (Rijkeboer & de Boo, 2010; Stallard, 2007). This is also supported by findings that indicate that other cognitive vulnerabilities are still under construction in youth (Cole et al., 2008; LaGrange et al., 2008). For instance, the nature of attributional style changes qualitatively over the course of middle childhood and early adolescence, and such cognitions do not begin to show stylelike characteristics until early adolescence (Cole et al., 2008). In adolescence, peer relationships become a primary source of intimacy, self-disclosure, and nurturance (McCarthy & Lumley, 2012). Therefore, experiences of victimization by peers can also contribute to the construction of dysfunctional schemas of oneself and of social relationships. In accordance with the increased role of peer relationships, a number of studies have found that victims of bullying experience profound changes in their cognitions as a consequence of victimization. For instance, a few studies have obtained support for the hypothesis that bullying victimization predicts other cognitive vulnerabilities, such as negative inference style (e.g., Gibb & Abela, 2008; Gibb, Stone, & Crossett, 2012; Mezulis, Hyde, & Abramson, 2006) and hopelessness (Hamilton et al., 2013). In contrast, research on bullying victimization’s impact on EMSs is scarce and is limited to cross-sectional studies. For example, McCarthy and Lumley (2012) explored the roles of

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

737

three sources of maltreatment (parents, peers, and intimate partners). They found that maltreatment by parents was more strongly associated with EMSs than was maltreatment perpetrated by other people (peers and intimate partners). In another study, adolescents who had been victimized by peers scored higher on the mistrust schema, which belongs to the disconnection/rejection domain, than did adolescents who had not been victimized (Estévez, Villardón, Calvete, Padilla, & Orue, 2010). Temperament as a moderator of the influence of maltreatment on EMSs As previously mentioned, Young’s model also involves the hypothesis that temperament interacts with environment in the development of EMSs. Namely, the model proposes that the interaction between temperament and maltreatment results in the frustration of the child’s basic needs (Young et al., 2003). In this context, two temperament traits, neuroticism, or negative affectivity, and low extraversion, or low positive affectivity, may play a relevant role due to their influence in the development of psychological disorders (Anderson, Veed, Inderbitzen-Nolan, & Hansen, 2010; Chorpita, Plummer, & Moffitt, 2000; for a review, see Epkins & Heckler, 2011). For instance, two children might react very differently to parental maltreatment. A child with low extraversion could withdraw and hide from the social world, thereby developing schemas within the disconnection/rejection domain, whereas a child with high extraversion could venture outwards and seek other, more positive, relationships, thereby developing more functional schemas. Overview of the present study The review of the available research indicates that no study has yet examined the conjoint influence of maltreatment experiences and temperament in the worsening of EMSs, as hypothesized by the schema therapy model. The present study aims to examine whether temperament and emotional abuse perpetrated by parents and peers, alone and interactively, contribute to the change of EMSs in adolescents. If so, this finding would be consistent with the available evidence indicating that maltreatment and temperament act as risk factors for the development of depression and social anxiety (Epkins & Heckler, 2011). Accordingly, the present study proposes that experiences of emotional abuse perpetrated by parents and peers interact with adolescents’ temperaments to predict the worsening of EMSs and that these EMSs in turn predict the increase in depressive and social anxiety symptoms over time. Examining evidence relevant to this hypothesis is important because it would contribute to understanding some mechanisms through which maltreatment leads to the development of depressive and social anxiety symptoms, two of the most prevalent emotional problems in adolescents (e.g., Esbjørn, Hoeyer, Dyrborg, Leth, & Kendall, 2010), which occur with substantial comorbidity (Essau, Conradt, & Petermann, 1999). The assessment of the mediational hypothesis proposed in this study requires taking measurements in at least in three waves to examine whether experiences of emotional abuse and temperament at Time 1 (T1) predict the worsening of EMSs at Time 2 (T2) and whether EMSs, in turn, predict an increase in depressive and social anxiety symptoms at Time 3 (T3). Specifically, this study included measures of two temperament dimensions (neuroticism and extraversion), two types of maltreatment experiences (emotional abuse by parents and peers), three schema domains (disconnection/rejection, impaired autonomy, and other-directedness), and depressive and social anxiety symptoms. Method Participants The initial sample included 1,281 adolescents aged 13–17 years (593 girls and 688 boys) who were high school students from 51 classrooms at eight educational centers in Bizkaia, Spain. The assessments occurred at three time points, each of which was separated by 6 months: at the beginning of the school year (T1), 6 months later (T2), and 1 year later (T3). Twohundred twenty-nine adolescents did not complete the measurements at one of the time points; their lack of participation was caused almost entirely by sickness or absence. The attrition rate also included participants who did not respond to some of the questionnaires and were therefore eliminated from the study. Consequently, the final sample consisted of 1,052 adolescents (499 girls and 553 boys) who completed the measures during all three waves of the study (participation rate: 82.12%). The final sample had a mean age of 13.61 years (SD = 1.41) at the beginning of the study. A series of t tests was conducted to examine differences in all study variables at T1 among the 1,052 adolescents who completed the three waves and those who failed to complete the study. None of these analyses were significant. Socioeconomic levels were determined by applying the criteria recommended by the Spanish Society of Epidemiology and Family and Community Medicine (2000) and from information provided about parental education and income. The distribution of the students among the socioeconomic levels was as follows: 19.1% low, 17.5% low–medium, 25.8% medium, 18.7% high–medium, and 18.6% high. Measures EMSs were assessed using the Young Schema Questionnaire-3 (YSQ-3; Young, 2006). The YSQ-3 consists of 90 items and assesses 18 cognitive schemas (five items per schema). The participants rated the items using a 6-point scale ranging

738

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

from 1 (completely untrue of me) to 6 (describes me perfectly). In this study, the YSQ-3 was used to assess the domains of disconnection/rejection, impaired autonomy, and other-directedness. The Spanish version of the YSQ-3 has shown good psychometric properties, with confirmation of the factor structure and adequate alpha coefficients for the scales (Calvete, Orue, & González-Díez, 2013). Whereas the majority of the studies have confirmed the EMSs factor structure of the YSQ, there have been mixed findings regarding the factor structure for the schema domains (for review, see Calvete et al., 2013). A satisfactory three-domain structure with a lower number of schemas has been obtained in a recent study (Calvete, Orue, Cámara, & Hankin, 2013) and was used in the current study. The domain of disconnection/rejection included the schemas of abandonment, mistrust, emotional deprivation, and defectiveness. The domain of impaired autonomy was represented by the schemas of vulnerability to harm and failure. The schemas within the domain of other-directedness were subjugation and need for acceptance. Thus, the measures of schema domains did not include the schemas of social isolation, self-sacrifice, enmeshment, and dependence, which, overall, are not particularly relevant for social anxiety and depression (e.g., Calvete et al., 2013; Calvete, Estevez, Lopez de Arroyabe, & Ruiz, 2005). The alpha coefficients were .89, .81, and .86 at T1 and .91, .84, and .88 at T2 for the disconnection/rejection, impaired autonomy, and other-directedness domains, respectively. Emotional abuse by parents was measured using a Spanish adaptation of the Psychological Abuse Scale of the Conflict Tactics Scales-Parent-to-Child version (CTS-PC, Straus, Hamby, Finkelhor, Moore, & Runyan, 1998). This scale consists of 5 items that are answered twice, once regarding the mother and again regarding the father (e.g., At home, my mother/my father makes me feel worthless). A back-translation method was used to elaborate the Spanish version of the scale. In the Spanish version participants indicate the frequency with which these experiences have occurred on a 4-point response scale ranging from 1 (Has not occurred) to 4 (Occurs very often). As the Spanish version of scale was developed in this study, its measurement model was examined and is presented in the “Results” section. The alpha coefficient was .78. Abuse by peers was measured using the Victimization Scale of the Peer Relations Questionnaire for Children (PRQ; Rigby, 1996). The Victimization Scale includes 6 items (e.g., At school they pick on me). Participants indicate the frequency with which this type of experience has occurred on a 4-point response scale ranging from 1 (Has not occurred) to 4 (Occurs very often). Previous research has provided support for the validity and reliability of this measure (Bond, Wolfe, Tollit, Butler, & Patton, 2007). In this study, only items describing emotional victimization were used. A back-translation method was used to elaborate the Spanish version of the scale. The structure of the scale was examined and is presented in the “Results” section. In this study, the alpha coefficient was .84. Temperament was measured using the Big Five Questionnaire-Children (Barbaranelli, Carpara, Rabasca, & Pastorelli, 2003). The Spanish version of the BFQ-C has displayed excellent psychometric properties, including test–retest reliability, internal consistency, and confirmation of its factor structure (Carrasco, Holgado, & Del Barrio 2005). In this study, two temperament dimensions were assessed: Neuroticism and extraversion. Neuroticism (11 items) refers to the tendency to experience or react with emotions such as fear, sadness, and anger. Extraversion (10 items) involves social engagement and reacting with positive emotions, such as joy and eagerness. Items are completed using a 5-point response score, from 5 (almost always) to 1 (almost never). The scores were recoded so that high scores indicated high neuroticism and extraversion. In this study, the alpha coefficients were .74 and .84 for extraversion and neuroticism, respectively. Depressive symptoms were assessed using the Center for Epidemiological Studies Depression Scale (CES-D; Radloff, 1977). The CES-D consists of 20 statements rated on a 4-point scale ranging from 0 (rarely or none of the time) to 3 (most or all of the ˜ time). Previous research with the Spanish version of the CES-D has confirmed its factorial structure (Calvete & Cardenoso, 1999). Mild depressed mood is indicated by a score of 16–22, moderate depressed mood is indicated by a score of 23–27, and scores of 28 or higher indicate severe depressed mood consistent with major depressive disorder (Radloff, 1991). In this study, the alpha coefficient at T1 was .85; at T2 and T3, it was .88. Social anxiety symptoms were assessed using the Social Anxiety Scale for Adolescents (SAS-A, La Greca & López, 1998). The SAS-A contains 18 items in the form of statements about oneself (e.g., I get nervous when I talk to peers that I don t know very well) that are rated on a 5-point scale ranging from 1 (not at all) to 5 (all the time). The SAS-A includes items regarding fear of negative evaluation, social avoidance and distress specific to new situations, and general social avoidance. The Spanish version of the questionnaire has demonstrated good psychometric properties (Olivares et al., 2005). The alpha coefficients in this study were 0.89 for T1 and 0.92 for both T2 and T3.

Procedure Data were collected at three measurement periods spaced 6 months apart. The participants completed measures of temperament, emotional abuse by parents and peers at T1, measures of EMSs at T1 and T2, and measures of depressive and anxiety symptoms at T1, T2, and T3. The Ethics Committee of the University of Deusto approved this study. Responses were kept anonymous to promote honesty, and participation was voluntary. Because there were no student names included on the surveys, the school staff chose to collect passive consent from parents. Thus, parents were notified and given the option of refusing to allow their child’s participation in the three waves of the study. None of the parents refused to allow their child to participate. All adolescents consented to participate. The adolescents completed the questionnaires in their classrooms. To match the questionnaire responses at T1, T2, and T3, a code known only by the participant was used. Some questionnaires could not be paired because of errors in the codes and were therefore eliminated; such cases comprised a portion of the attrition rate. The questionnaires took 45–60 min to complete.

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

739

Results Preliminary analyses: psychometric properties of the emotional abuse measures Because the measures of emotional abuse were translated into Spanish for this study, their structure was examined by confirmatory factor analysis with LISREL 8.8 (Jöreskog and Sörbom, 2006). The models were tested using weighted leastsquares estimation (WLS). Following the recommendations of several authors (Hu & Bentler, 1999), the goodness of fit was evaluated using the comparative fit index (CFI), the nonnormative fit index (NNFI), and the root mean square error of approximation (RMSEA). Generally, CFI and NNFI values of .90 or higher reflect a good fit. In addition, RMSEA values lower than .06 indicate excellent fit, whereas values between .06 and .08 indicate acceptable fit. I hypothesized that the items from the Emotional Abuse Scale of the CTS-PC could be explained by a one-factor structure. This measurement model showed excellent fit indexes, 2 (5, n = 1,052) = 22, RMSEA = .058 (90% Confidence Interval – CI: .043; .070), NNFI = .95, CFI = .97. The factor loadings ranged between .67 and .89. Similarly, a one-factor model was hypothesized for the Victimization Scale of the PRQ. The measurement model showed good fit indexes, 2 (9, n = 1,052) = 46, RMSEA = .063, 90% CI [.048, .079], NNFI = .98, CFI = .99. Factor loadings ranged between .64 and .95. Descriptive statistics and correlation between variables Table 1 displays the descriptive statistics and correlations between the variables of the study. Most of the variables were significantly correlated. For instance emotional abuse by peers was associated with EMSs. However, extraversion was not associated with neuroticism, parental emotional abuse, or several EMSs. Emotional symptoms were significantly associated with most of the variables of the study. The overall prevalence of clinically significant symptoms of social anxiety, based on the SAS-A cutoff score of 50 (La Greca & López, 1998), was 25.8%, 25.4%, and 23.8% at T1, T2, and T3, respectively. The prevalence of severe depressed mood (cutoff score > 28) was 13.3%, 11.6%, and 11.2% at T1, T2, and T3. Rates of comorbid depression and social anxiety were 7.1, 7.1, and 6.8% at T1, T2, and T3, respectively. Emotional abuse, temperament, EMSs, and depressive and social anxiety symptoms: mediational model Establishing measurement invariance over time. Structural Equation Modeling was used to test the hypothesis of the study. All of the models were tested via maximum likelihood (ML) estimation using LISREL 8.8 (Jöreskog & Sörbom, 2006). In the first step, the measurement model’s fit for the latent variables of the study was examined. Item-parcels were used as indicators of the latent variables. Two parcels were used as indicators for the schema domains, extraversion, neuroticism, emotional abuse by parents and peers, and depressive and social symptoms. To represent the latent interactions (i.e., Neuroticism × Parental Emotional Abuse; Neuroticism × Emotional Abuse by Peers; Extraversion × Parental Emotional Abuse; and Extraversion × Emotional Abuse by Peers), the residual centering procedure described by Little, Bovaird, and Widaman (2006) was used. This procedure indicates that, to create orthogonalized indicators for a latent two-order interaction construct, each possible product term from two sets of indicators for two latent constructs (e.g., neuroticism and emotional abuse by peers) must be formed. As in this study, two indicators were used for each main latent variable, meaning that four uncentered product terms were computed (e.g., Neuroticism Parcel 1 × Emotional Abuse by Peers Parcel 1). Each of these resulting product terms was then individually regressed onto the first-order effect indicators of the constructs. The residuals for this regression were saved and used as an indicator of the product term. The procedure was repeated for each of the four product terms. Finally, the four orthogonalized product terms were used as indicators of the latent interaction construct. In addition, the latent interaction construct was not allowed to correlate with the main-effect latent variables. The error terms of the same variable assessed on different occasions were conceptualized as correlated based on the assumption that factors contributing to measurement error in any specific variable will be consistent across measurement occasions. Fig. 1 displays all the latent variables of the model. To determine whether the measurement model could be considered equivalent (or invariant) over time, the model fit of two alternative models (differing in levels of parameter restrictions) was tested and compared (Cole & Maxwell, 2003) using the corrected chi-squared difference test (Satorra & Bentler, 2001). First, a longitudinal confirmatory factor analysis was undertaken, which included all observed and latent variables from each time with freely estimated parameters. The measurement model consisted of 20 latent variables corresponding to the two temperament traits (neuroticism and extraversion), parental emotional abuse, emotional abuse by peers, four interactive terms (Neuroticism × Parental Emotional Abuse; Neuroticism × Emotional Abuse by Peers; Extraversion × Parental Emotional Abuse; and Extraversion × Emotional Abuse by Peers), three schema domains (disconnection/rejection, impaired autonomy, and other-directedness) at T1 and T2, and depressive and social anxiety symptoms at T1, T2, and T3. This measurement model showed good fit indexes, 2 (959, n = 1,052) = 2,781, p < .001, RMSEA = .041, 90% CI [.040, .044], NNFI = .98, CFI = .98. This model was compared with a more restrictive model, in which factor loadings within constructs across time points were specified as equal: 2 (973, n = 1,052) = 2,802, p < .001, RMSEA = .042, 90% CI [.041, .044], NNFI = .99, CFI = .99. This comparison used the corrected chisquared difference test, as applied by Crawford and Henry (2003), and showed that the fit of the more-restrictive longitudinal model was not significantly worse than that of the less-restrictive longitudinal model (specifying freely estimated parameters): 2 (14, n = 1,052) = 13.41, p = .49.

740

Table 1 Descriptive statistics and correlations between variables. 2

1 .17** .21** −.06 .32** .29** .27** .28** .27** .24** .31** .27** .25** .17** .18** .17**

1 .27** .07* .42** .31** .37** .37** .30** .33** .32** .28** .25** .36** .33** .29**

1 −.06 .52** .39** .48** .46** .38** .43** .48** .43** .43** .49** .40** .37**

1 .12** .03 −.01 −.10** −.04 −.04 −.05 −.10** −.15** −.16** −.16** −.16**

Mean Standard deviation Range

9.05 3.29 6–24

7.82 2.66 6–24

24.43 7.74 11–55

20.73 6.07 10–50

* **

p < .05. p < .001.

3

4

5

6

7

8

9

10

1 .64** .51** .66** .45** .53** .43** .40** .49** .38** .36**

1 .60** .48** .67** .55** .44** .38** .70** .54** .47**

1 .73** .82** .48** .65** .53** .56** .71** .55**

1 .65** .44** .56** .47** .43** .55** .44**

1 .45** .56** .46** .56** .73** .57**

2.48 0.94 1–6

2.88 0.93 1–6

2.21 0.84 1–6

2.32 0.94 1–6

2.74 0.97 1–6

11

12

13

1 .58** .53** .47** .35** .35**

1 .67** .39** .49** .45**

1 .37** .45** .55**

14

15

16

1 .70** .80** .68** .54** .59** .63** .51** .47** .67** .52** .47** 2.32 0.82 1–6

17.64 9.04 0–60

16.24 9.62 0–60

25.86 9.44 0–60

1 .61** .55**

1 .66**

41.53 12.73 18–90

40.56 13.36 18–90

1 39.85 13.55 18–90

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

1 1. Parental emotional abuse 2. Emotional abuse by peers 3. Neuroticism 4. Extraversion 5. T1 Disconnection/rejection 6. T1 Impaired autonomy 7. T1 Other-directedness 8. T2 Disconnection/rejection 9. T2 Impaired autonomy 10. T2 Other-directedness 11. T1 Depression 12. T2 Depression 13. T3 Depression 14. T1 Social anxiety 15. T2 Social anxiety 16. T3 Social anxiety

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

T1 Dep

T2 Dep

T1 D/R

T2 D/R

741

T3 Dep

N E T1 OD

T2 OD

EB

PEA T1 IA

T2 IA

N x EB N x EPA E x EB E x PEA T1 Anx

T2 Anx

T3 Anx

Fig. 1. Full model linking emotional abuse, temperament, schema domains, and symptoms of depression and social anxiety. Note: Dep = depressive symptoms, Anx = anxiety symptoms, D/R = disconnection/rejection, N = neuroticism, E = extraversion, OD = other-directedness, IA = impaired autonomy, PEA = parental emotional abuse, EB = emotional bullying.

Autoregressive model. The next step was to test an autoregressive model, which included auto-regressive paths between T1, T2, and T3 measures of depressive and social anxiety symptoms and between T1 and T2 measures of schema domains. The coefficients for the T1–T2 autoregressive paths were .52, .51, .56, .68, and .72 for disconnection/rejection, impaired autonomy, other-directedness, and depressive and social anxiety symptoms, respectively. The coefficients for the T2–T3 autoregressive paths were .47 and .50 for depressive and social anxiety symptoms. Fit indices were excellent for the autoregressive model, 2 (987, n = 1,052) = 3,709, p < .001; RMSEA = .051, 90% CI [.049, .053], NNFI = .98, CFI = .99. Hypothesized mediational model. Building on the above autoregressive model, the fit of the hypothesized model was tested. It included paths (a) from T1 abuse variables, extraversion, neuroticism, and interactive terms to T2 schema domains, and (b) from T2 schema domains to T3 depressive and social anxiety symptoms. In addition, direct paths from T1 variables to T3 symptoms were estimated. Fit indices were excellent for this model, 2 (974, n = 1,052) = 3,540, p < .001; RMSEA = .05, 90% CI [.048, .052], NNFI = .98, CFI = .99. The results indicated that both emotional abuse by parents and peers were significantly associated with all schema domains at T1 (all coefficients ranging between .25 and .41). Regarding predictive associations, both neuroticism and emotional abuse by peers significantly predicted the residual increase in all schema domains at T2. Furthermore, T2 disconnection/rejection predicted the increase in depressive symptoms at T3, and T2 other-directedness predicted the increase in social anxiety symptoms at T3. The model also showed significant direct paths from neuroticism and extraversion to T3 depressive and social anxiety symptoms, and from abuse by parents to T3 depressive symptoms. No interaction term between emotional abuse and temperament predicted schema domains or symptoms. A more parsimonious model was estimated that included only significant paths. The fit indices were excellent for this model, 2 (798, n = 1,052) = 2,537, p < .001; RMSEA = .04, 90% CI [.04, .04], NNFI = .99, CFI = .99. Fig. 2 displays the resulting mediational model. This model explained 42%, 40%, and 47%, respectively, of the variance of disconnection/rejection, impaired autonomy, and other-directedness domains, respectively, and 43% and 51% of depressive and social anxiety symptoms, respectively. The

742

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

0.21(0.04)**

0.12(0.05)*

Neuroticism

0.14(0.04)**

0.18(0.03)** 0.17(0.05)** Abuse by parents

0.10(0.03)**

0.17(0.04)** 0.13(0.04)**

Abuse by peers

T2 Disconnection

T2 Autonomy

0.17(0.04)** 0.13(0.04)**

T3 Depressive symptoms

T2 Otherdirectedness

0.19(0.03)**

T3 Social anxiety

-0.17(0.03)**

-0.06(0.03)*

Extraversion

Fig. 2. The mediational model between emotional abuse, temperament, schemas, and emotional symptoms. Note. In this model, all of the T2 and T3 variables include previous measurements that are not represented for the sake of clarity. The values given are nonstandardized coefficients. Standard errors are in parentheses. *p < .0 and **p < .001.

significance of the model’s paths was tested via a bootstrapping procedure. This procedure was particularly important for testing the indirect effects of neuroticism and bullying on T3 symptoms via EMSs. In this study, 1,000 bootstrap samples were created from the original data set by random sampling with replacement, and the covariance matrix was estimated from each sample. The second step was to conduct the structural model 1,000 times with these 1,000 bootstrap covariances to yield 1,000 estimations of each path coefficient. The third step was to use LISREL’s saved output of the 1,000 estimations of each path coefficient to calculate an estimate of the indirect effect. The final step was to determine whether the 95% CI for the estimated indirect effect included zero. According to Shrout and Bolger’s (2002) proposal, an indirect effect is significant at the .05 level if the 95% confidence level does not include zero. Table 2 shows the results. Using Shrout and Bolger’s criterion, the results indicated that emotional abuse by peers and neuroticism predicted the increase of depressive symptoms via disconnection/rejection schemas and the increase of social anxiety via other-directedness schemas.

Discussion The purpose of this study was twofold: (a) to evaluate the role of emotional abuse and temperament in the construction of EMSs in adolescents and (b) to test whether changes in EMSs act as paths connecting experiences of emotional abuse and temperament with symptoms of depression and social anxiety. The study’s results are summarized below.

Table 2 Confidence intervals for the indirect and direct effects of emotional abuse and temperament by bootstrapping.

Effects on depressive symptoms Emotional bullying → disconnection/rejection → depressive symptoms Neuroticism → disconnection/rejection → depressive symptoms Neuroticism → depressive symptoms Total effect of neuroticism on depressive symptoms Extraversion → depressive symptoms Parental emotional abuse → depressive symptoms Effects on social anxiety Emotional bullying → other-directedness → social anxiety Neuroticism → other-directedness → social anxiety Neuroticism → social anxiety Total effect of Neuroticism on social anxiety Extraversion → social anxiety Note: 1,000 bootstrap samples. BC, bias corrected; CI, confidence intervals.

Mean

BC bootstrap 95% CI

.015 .020 .216 .236 −.171 .113

[.015, .016] [.019, .021] [.213, .219] [.232, .240] [−.174, −.169] [.111, .116]

.030 .036 .168 .204 −.083

[.029, .031] [.035, .037] [.165, .171] [.200, .208] [−.080, −.083]

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

743

The role of emotional abuse and temperament in the development of EMSs Regarding the first objective, the results indicate that victimization by peers and neuroticism are independent predictors of the change in the three schema domains assessed in this study. That is, both the experiences of emotional abuse by peers and neuroticism predict a worsening of schemas in the domains of disconnection/rejection, impaired autonomy and performance, and other-directedness. These results are consistent with those obtained in previous cross-sectional studies that showed associations between neuroticism and numerous EMSs (e.g., Muris, 2006; Sava, 2009; Thimm, 2010b) and with previous evidence of predictive associations between temperament and other cognitive vulnerabilities (Mezulis et al., 2006). Findings indicate that the emotional abuse perpetrated by peers, not the emotional abuse perpetrated by parents, predicts the worsening of the schemas. The predictive association between bullying victimization and schemas is of great importance because it suggests that peer victimization experienced in adolescence is crucial to the development of dysfunctional schemas. In particular, as a consequence of being bullied, adolescents may develop schemas involving thoughts and feelings of being rejected and abused by others (disconnection/rejection), feelings of being defective to some extent (disconnection/rejection), a sense of vulnerability and failure (impaired autonomy), and the need to satisfy the desires of others in an attempt to gain acceptance (other-directedness) and eventually to avoid bullying. Nevertheless, although emotional abuse perpetrated by parents did not predict the worsening of EMSs over time, it could have influenced the origin of EMSs. In fact, it was concurrently associated with measures of EMSs at T1. This is consistent with findings from previous studies that showed significant correlations between parental abuse and dysfunctional parenting styles and EMSs (Carr & Francis, 2010; McCarthy & Lumley, 2012; Thimm, 2010a; Wright et al., 2009). Thus, it is possible that the negative influence of parental emotional abuse in the development of EMSs had already occurred in childhood prior to this study. As a consequence, not much further change in EMSs was predicted by parental emotional abuse experienced in adolescence. In this study, emotional abuse by parents and peers appear to be fairly distinct and reveal only a low correlation (.17), highlighting the importance of including various types of maltreatment in studies. As mentioned above, only one previous study compared the role of various sources of maltreatment in EMSs (McCarthy & Lumley, 2012). In their cross-sectional study, these authors found that abuse perpetrated by parents was associated more strongly with EMSs than abuse perpetrated by others, including both bullying victimization and partner violence. Moreover, their study was conducted with undergraduate students, whereas the present study was conducted with adolescents for whom experiences of maltreatment were more recent. Therefore, there are many differences between the previous study and the present one, which can explain differences in their results. In any case, the influence of bullying victimization on the worsening of EMSs is congruent with the importance of peer relationships in adolescence (Lumley & Harkness, 2007). Although emotional abuse by parents did not predict the worsening of EMSs, it was directly associated with depressive symptoms, both concurrently and prospectively. This result is consistent with many previous studies that showed that parental maltreatment is a predictor of the onset of depression (Hankin, 2005; Liu, Alloy, Abramson, Iacoviello, & Whitehouse, 2009). Nevertheless, the influence of parental emotional abuse on depressive symptoms was not mediated by a change in EMSs. The influence of parental emotional abuse on EMSs could have occurred early in childhood so that no additional change in EMSs was observed during the time when the study was performed. Alternatively, the influence of parental emotional abuse on depressive symptoms could occur through the development of other cognitive vulnerabilities, such as negative inferences and ruminative responses (Gibb & Abela, 2008; Gibb et al., 2006; Padilla & Calvete, in press), which were not measured in this study. Contrary to expectations, there was no significant interaction between temperament dimensions and emotional abuse (perpetrated by parents and peers) in the prediction of EMSs and/or symptoms of depression and social anxiety. Nevertheless, as mentioned before, as this study did not include measures of early emotional abuse that could have occurred prior to this investigation, interactions between early experiences and temperament could not be tested properly. Furthermore, this study examined only one component of the theory, which concerns the role of emotional abuse. The schema therapy model (Young et al., 2003) suggests that many other early experiences can contribute to the genesis of EMSs. One of these alternative circumstances occurs when the child experiences deprivations and acquires such schemas as emotional deprivation or abandonment through deficits in stability, understanding, or love in early environments. Another type of life experience that creates schemas is selective identification with a parent’s thoughts, feelings, experiences, and behaviors (Young et al., 2003). For instance, such identification could take place in situations of violence witnessing. The role of these experiences in the development of EMSs should be evaluated in future longitudinal studies. Contributions to an explanation of the development of depressive and social anxiety symptoms in adolescents The second objective was to assess whether the influence of emotional abuse experiences and temperament in depression and social anxiety is mediated by EMSs, thereby identifying some of the mechanisms that contribute to developing depression and social anxiety. The results confirmed the mediational hypothesis for both experiences of emotional bullying victimization and neuroticism. In the case of depression, mediation occurred through the disconnection/rejection domain, whereas in the case of social anxiety, it occurred through the other-directedness domain. These results are consistent with previous evidence about the role of these schema domains in depressive symptoms (Calvete et al., 2013a; Eberhart, Auerbach, Bigda-Peyton, & Abela, 2011; Lumley & Harkness, 2007; Roelofs, Lee, Ruijten, & Lobbestael, 2011) and social anxiety (e.g., Calvete et al.,

744

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

2013b; Hinrichsen et al., 2004). Furthermore, the mediational model tested in this study expands the findings of previous cross-sectional studies that examined EMSs as mediators between experiences of maltreatment and psychological disorders (e.g., Calvete & Orue, 2013; Carr & Francis, 2010; Wright et al., 2009). The results also indicate some common prospective relationships between emotional abuse, temperament, and emotional symptoms. The fact that some of the predictors of depression and social anxiety are common may partly explain the comorbidity between the two disorders. In both cases, emotional bullying victimization, neuroticism, and low extraversion predicted the increase of symptoms over time. Differences emerged mainly in two aspects: cognitive and emotional mediators and parental maltreatment. Whereas EMSs in the disconnection/rejection domain predicted an increase in depressive symptoms, EMSs in the other-directedness domain predicted the increase in social anxiety. The second difference lay in the role of parental emotional abuse, which predicted an increase in depressive symptoms but not in social anxiety. This suggests that emotional abuse within the family is a specific risk factor for depression. Strengths and limitations of the study This study has some limitations that provide opportunities and challenges for future research. The first limitation is that the measures used were based exclusively on self-reports, and this may cause increased associations between measures. Furthermore, the study lacked a historical measure of parental maltreatment. Future research should examine rigorously the role of early parental abuse by means of structured interviews such as the adolescent version of the Childhood Experience of Care and Abuse (Bifulco, Brown, & Harris, 1994). This would allow assessing the role of early parental abuse in the origin of EMSs. The second limitation is that the sample consisted of a nonclinical sample of adolescents, and replication in clinical samples is required. Nevertheless, clinical samples are not without problems and both types of studies (clinical vs. nonclinical studies) likely complement each other. Moreover, depressive and social anxiety symptoms appear to be recognized risk factors for the development of depression and social anxiety disorders, and symptoms generally show the same correlates as diagnosed depression and social anxiety disorders (Epkins & Heckler, 2011). The third limitation is that the study did not assess bidirectional relationships among some variables (e.g., between temperament and maltreatment). Although temperament traits are considered to be genetically based core dimensions, neuroticism has been found to change over time, as a result of maltreatment (DeYoung, Cicchetti, & Rogosch, 2011). Furthermore, recent findings on stress generation suggest that emotional symptoms, EMSs and temperament traits can generate adversity (Calvete et al., 2013a; Gibb & Hanley, 2010) and that these variables could increase the risk of bullying victimization. Finally, this study did not include measures of stressors, which, according to the diathesis stress models, prime schemas and lead to emotional problems. Thus, a low activation of the EMSs could have explained the relatively weak paths connecting T2 EMSs with T3 symptoms. Despite these limitations, the study also has several strengths, such as the use of a large sample of adolescents that was followed over three waves. The study addresses the etiology of two emotional problems that are highly prevalent in adolescence and that often are comorbid. Moreover, the fact that both psychological problems, along with measures of temperament, EMSs, and emotional abuse, were integrated in the same structural model helps to elucidate the common and specific etiological factors for depression and social anxiety. Most of the previous studies on the role of emotional abuse in the development of cognitive vulnerabilities ignored the potential influence of temperament. Although the results of this study indicate that both factors act independently in the genesis of EMSs, future studies should evaluate the hypothesis of the interaction for other early life experiences (e.g., neglect, witnessing violence) and in clinical samples of adolescents diagnosed with depression or social anxiety. Finally, the findings suggest that EMSs are still under construction in adolescence and that they are affected by relevant experiences that occur at that age, such as emotional bullying victimization. This implies that childhood and adolescence constitute a promising time for early interventions focusing on dysfunctional schemas, especially in those children and adolescents who have experienced victimization. The schema therapy model (Young et al., 2003) includes several strategies to assess and to modify EMSs, such as emotional imagery, interpersonal techniques, cognitive restructuring, and self-empowerment exercises, which could be used with adolescents who have experienced emotional abuse. Conclusion In conclusion, the study indicates that both emotional abuse by peers and temperament (i.e., neuroticism) predict a worsening of EMSs, which constitutes one of the mechanisms through which these factors increase the risk of depression and social anxiety. The influence of both risk factors (i.e., emotional abuse and temperament) seem to be independent each of the other, although future research should examine whether temperament interacts with other adversities, such as child neglect and violence witnessing, in the genesis of emotional and cognitive schemas. References Alloy, L. B., Abramson, L. Y., Smith, J. M., Gibb, B. E., & Neeren, A. M. (2006). Role of parenting and maltreatment histories in unipolar and bipolar mood disorders: Mediation by cognitive vulnerability to depression. Clinical Child and Family Psychology Review, 9, 23–64.

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

745

Anderson, E. R., Veed, G. J., Inderbitzen-Nolan, H. M., & Hansen, D. J. (2010). An evaluation of the applicability of the tripartite constructs to social anxiety in adolescents. Journal of Clinical Child and Adolescent Psychology, 39, 195–207. http://dx.doi.org/10.1080/15374410903532643 Barbaranelli, C., Carpara, G., Rabasca, A., & Pastorelli, C. (2003). A questionnaire for measuring the Big Five in late Childhood. Personality and Individual Differences, 34, 645–664. Bifulco, A., Brown, G. W., & Harris, T. O. (1994). Childhood experience of care and abuse (CECA): A retrospective interview measure. Journal of Child Psychology and Psychiatry and Allied Disciplines, 35, 1419–1435. Bond, L., Wolfe, S., Tollit, M., Butler, H., & Patton, G. (2007). A comparison of the gatehouse bullying scale and the peer relations questionnaire for students in secondary school. Journal of School Health, 77, 75–79. http://dx.doi.org/10.1111/j.1746-1561.2007.00170.x Bruce, L. C., Heimberg, R. G., Blanco, C., Schneier, F. R., & Liebowitz, M. R. (2012). Childhood maltreatment and social anxiety disorder: Implications for symptom severity and response to pharmacotherapy. Depression and Anxiety, 29, 131–138. http://dx.doi.org/10.1002/da.20909 ˜ Calvete, E., & Cardenoso, O. (1999). Creencias y síntomas depresivos: Resultados preliminares en el desarrollo de una Escala de Creencias Irracionales abreviada [Beliefs and depressive symptoms: Preliminary results in the development of a brief irrational beliefs scale]. Anales de Psicología, 15, 179–190. Calvete, E., & Orue, I. (2013). Cognitive mechanisms of the transmission of violence: Exploring gender differences among adolescents exposed to family violence. Journal of Family Violence, 28, 73–84. http://dx.doi.org/10.1007/s10896-012-9472-y Calvete, E., Estevez, A., Lopez de Arroyabe, E., & Ruiz, P. (2005). The Schema Questionnaire—Short form: Structure and relationship with automatic thoughts and symptoms of affective disorders. European Journal of Psychological Assessment, 21, 91–100. Calvete, E., Orue, I., Cámara, M., & Hankin, B. L. (2013). A longitudinal test of the vulnerability-stress model with early maladaptive schemas for depressive and social anxiety symptoms in adolescents. (submitted for publication). Calvete, E., Orue, I., & Hankin, B. L. (2013a). Transactional relationships among cognitive vulnerabilities, stressors, and depressive symptoms in adolescence. Journal of Abnormal Child Psychology, 41, 399–410. http://dx.doi.org/10.1007/s10802-012-9691-y Calvete, E., Orue, I., & Hankin, B. L. (2013b). Early maladaptive schemas and social anxiety in adolescents: The mediating role of anxious automatic thought. Journal of Anxiety Disorders, 27, 278–288. Calvete, E., Orue, I., & González-Díez, Z. (2013). An examination of the structure and stability of early maladaptive schemas by means of the Young Schema Questionnaire-3. European Journal of Psychological Assessment, 29, 283–290. http://dx.doi.org/10.1027/1015-5759/a000158 Carr, S. N., & Francis, A. P. (2010). Early maladaptive schemas and personality disorder symptoms: An examination in a non-clinical sample. Cognitive Therapy and Research, 83(4), 333–349. http://dx.doi.org/10.1348/147608309X481351 ˜ Carrasco, M. A., Holgado, F. P., & Del Barrio, M. V. (2005). Dimensionalidad del cuestionario de los cinco grandes (BFQ-N) en población infantil espanola [Dimensionality of the Big Five Questionnaire (BFQ-N) in Spanish child population]. Psicothema, 17(2), 286–291. Chorpita, B. F., Plummer, C. M., & Moffitt, C. E. (2000). Relations of tripartite dimensions of emotion to childhood anxiety and mood disorders. Journal of Abnormal Child Psychology, 28, 299–310. http://dx.doi.org/10.1023/A:1005152505888 Cicchetti, D., & Valentino, K. (2006). An ecological transactional perspective on child maltreatment: Failure of the average expectable environment and its influence upon child development. In D. Cicchetti, & D. J. Cohen (Eds.), Developmental psychopathology (vol. 3) (2nd ed., vol. 3, pp. 129–201). New York, NY: Wiley. Cole, D. A., & Maxwell, S. E. (2003). Testing meditational models with longitudinal data: Questions and tips in the use of structural equation modeling. Journal of Abnormal Psychology, 112, 558–577. Cole, D. A., Ciesla, J. A., Dallaire, D. H., Jacquez, F. M., Pineda, A. Q., LaGrange, B., & Felton, J. W. (2008). Emergence of attributional style and its relation to depressive symptoms. Journal of Abnormal Psychology, 117, 16–31. Cole, D. A., Dukewich, T. L., Roeder, K., Sinclair, K. R., McMillan, J., Will, E., Bilsky, S. A., Martin, N. C., & Felton, J. W. (2013). Linking peer victimization to the development of depressive self-schemas in children and adolescents. Journal of Abnormal Child Psychology, 1–12. Crawford, J. R., & Henry, J. D. (2003). The Depression Anxiety Stress Scales: Normative data and latent structure in a large non-clinical sample. British Journal of Clinical Psychology, 42, 111–131. DeYoung, C. G., Cicchetti, D., & Rogosch, F. A. (2011). Moderation of the association between childhood maltreatment and neuroticism by the corticotropinreleasing hormone receptor 1 gene. Journal of Child Psychology and Psychiatry, 52, 898–906. http://dx.doi.org/10.1111/j.1469-7610.2011.02404.x Eberhart, N. K., Auerbach, R. P., Bigda-Peyton, J., & Abela, J. R. Z. (2011). Maladaptive schemas and depression: Tests of stress generation and diathesis-stress models. Journal of Social & Clinical Psychology, 30, 75–104. http://dx.doi.org/10.1521/jscp.2011.30.1.75 Epkins, C. C., & Heckler, D. R. (2011). Integrating etiological models of social anxiety and depression in youth: Evidence for a cumulative interpersonal risk model. Clinical Child and Family Psychology Review, 14, 329–376. http://dx.doi.org/10.1007/s10567-011-0101-8 Esbjørn, B., Hoeyer, M., Dyrborg, J., Leth, I., & Kendall, P. C. (2010). Prevalence and co-morbidity among anxiety disorders in a national cohort of psychiatrically referred children and adolescents. Journal of Anxiety Disorders, 24, 866–872. http://dx.doi.org/10.1016/j.janxdis.2010.06.009 Essau, C. A., Conradt, J., & Petermann, F. (1999). Frequency and comorbidity of social phobia and social fears in adolescents. Behaviour Research and Therapy, 37, 831–843. Estévez, A., Villardón, L., Calvete, E., Padilla, P., & Orue, I. (2010). Adolescentes víctimas de cyberbullying: Prevalencia y características [Adolescent victims of cyberbullying: Prevalence and characteristics]. Behavioral Psychology, 18, 73–89. Gibb, B. E., & Abela, J. R. Z. (2008). Emotional abuse, peer victimization and the development of children’ negative inferential styles and depressive symptoms. Cognitive Therapy and Research, 32, 161–176. http://dx.doi.org/10.1007/s10608-006-9106-x Gibb, B. E., & Hanley, A. J. (2010). Depression and interpersonal stress generation in children: Prospective impact on relational versus overt victimization. International Journal of Cognitive Therapy, 3, 358–367. http://dx.doi.org/10.1521/ijct.2010.3.4.358 Gibb, B. E., Butler, A. C., & Beck, J. S. (2003). Childhood abuse, depression, and anxiety in adult psychiatric outpatients. Depression and Anxiety, 17, 226–228. http://dx.doi.org/10.1002/da.10111 Gibb, B. E., Abramson, L. Y., & Alloy, L. B. (2004). Emotional maltreatment from parents, verbal peer victimization and cognitive vulnerability to depression. Cognitive Therapy and Research, 28, 1–21. Gibb, B. E., Alloy, L. B., Walshaw, P. D., Comer, J. S., Shen, G. H. C., & Villari, A. G. (2006). Predictors of attributional style change in children. Journal of Abnormal Child Psychology, 34, 425–439. http://dx.doi.org/10.1007/s10802-006-9022-2 Gibb, B. E., Chelminski, I., & Zimmerman, M. (2007). Childhood emotional, physical, and sexual abuse, and diagnoses of depressive and anxiety disorders in adult psychiatric outpatients. Depression and Anxiety, 24(4), 256–263. Gibb, B. E., Stone, L. B., & Crossett, S. E. (2012). Peer victimization and prospective changes in children’s inferential styles. Journal of Clinical Child & Adolescent Psychology, 41, 561–569. Hamilton, J. L., Shapero, B. G., Stange, J. P., Hamlat, E. J., Abramson, L. Y., & Alloy, L. B. (2013). Emotional maltreatment, peer victimization, and depressive versus anxiety symptoms during adolescence: Hopelessness as a mediator. Journal of Clinical Child & Adolescent Psychology, 42, 332–347. Hankin, B. L. (2005). Childhood maltreatment and psychopathology: Prospective test of attachment, cognitive vulnerability and stress as mediating process. Cognitive Therapy and Research, 29, 645–671. http://dx.doi.org/10.1007/s10608-005-9631-z Harkness, K. L., Bruce, A. E., & Lumley, M. N. (2006). The role of childhood abuse and neglect in the sensitization to stressful life events in adolescent depression. Journal of Abnormal Psychology, 115, 730–741. http://dx.doi.org/10.1037/0021-843X.115.4.730 Hinrichsen, H., Waller, G., & Emanuelli, F. (2004). Social anxiety and agoraphobia in the eating disorders: Associations with core beliefs. Journal of Nervous and Mental Disease, 192, 784–787. Hu, L., & Bentler, P. M. (1999). Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternatives. Structural Equation Modeling, 6, 1–55. http://dx.doi.org/10.1080/10705519909540118 Ingram, R. E. (2003). Origins of cognitive vulnerability to depression. Cognitive Therapy and Research, 27, 77–88. http://dx.doi.org/10.1023/A:1022590730752 Jöreskog, K. G., & Sörbom, D. (2006). LISREL 8.8 [Computer software]. Skokie, IL: Scientific Software International.

746

E. Calvete / Child Abuse & Neglect 38 (2014) 735–746

Kim, J., & Cicchetti, D. (2010). Longitudinal pathways linking child maltreatment, emotion regulation, peer relations, and psychopathology. Journal of Child Psychology and Psychiatry, 51, 706–716. http://dx.doi.org/10.1111/j.1469-7610.2009.02202.x La Greca, A. M., & López, N. (1998). Social anxiety among adolescents: Linkage with peer relations and friendships. Journal of Abnormal Child Psychology, 26, 83–94. LaGrange, B., Cole, D. A., Dallaire, D. H., Ciesla, J. A., Pineda, A. Q., Truss, A. E., & Folmer, A. (2008). Developmental changes in depressive cognitions: A longitudinal evaluation of the Cognitive Triad Inventory for Children. Psychological Assessment, 20, 217–226, doi:10.10371040-3590.20.3.217. Little, T. D., Bovaird, J. A., & Widaman, K. F. (2006). On the merits of orthogonalizing powered and product terms: Implications for modeling interactions among latent variables. Structural Equation Modeling, 13, 497–519. Liu, R. T., Alloy, L. B., Abramson, L. Y., Iacoviello, B. M., & Whitehouse, W. G. (2009). Emotional maltreatment and depression: prospective prediction of depressive episodes. Depression & Anxiety, 26, 174–181. http://dx.doi.org/10.1002/da.20545 Lumley, M. N., & Harkness, K. L. (2007). Specificity in the relations among childhood adversity, early maladaptive schemas, and symptom profiles in adolescent depression. Cognitive Therapy and Research, 31, 639–657. http://dx.doi.org/10.1007/s10608-006-9100-3 McCarthy, M. C., & Lumley, M. N. (2012). Sources of emotional maltreatment and the differential development of unconditional and conditional schemas. Cognitive Behaviour Therapy, 41, 288–297. http://dx.doi.org/10.1080/16506073.2012.676669 Mezulis, A., Hyde, J. S., & Abramson, L. Y. (2006). The developmental origins of cognitive vulnerability to depression: Temperament, parenting, and negative life events in childhood as contributors to negative cognitive style. Developmental Psychology, 42, 1012–1025. http://dx.doi.org/10.1037/0012-1649.42.6.1012 Muris, P. (2006). Maladaptive schemas in non-clinical adolescents: Relations to perceived parental rearing behaviours, big five personality factors and psychopathological symptoms. Clinical Psychology & Psychotherapy, 13, 405–413. http://dx.doi.org/10.1002/cpp.506 Olivares, J., Ruiz, J., Hidalgo, M. D., García-López, L. J., Rosa, A. I., & Piqueras, J. A. (2005). Social anxiety scale for adolescents (SAS-A): Psychometric properties in a Spanish-speaking population. International Journal of Clinical and Health Psychology, 5, 85–97. Padilla, P., & Calvete, E. (in press). Cognitive vulnerabilities as mediators between maltreatment and depressive symptoms. Journal of Abnormal Child Psychology. Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in general population. Applied Psychological Measurement, 1, 385–401. Radloff, L. S. (1991). The use of the Center for Epidemiological Studies Depression Scale in adolescents and young adults. Journal of Youth and Adolescence, 20, 149–166. Reijntjes, A., Kamphuis, J. H., Prinzie, P., & Telch, M. J. (2010). Peer victimization and internalizing problems in children: A metaanalysis of longitudinal studies. Child Abuse & Neglect, 34, 244–252. http://dx.doi.org/10.1016/j.chiabu.2009.07.009 Rigby, K. (1996). Manual for the Peer Relations Questionnaire (The PRQ). Underdale, South Australia: University of South Australia. Rijkeboer, M. M., & de Boo, G. M. (2010). Early maladaptive schemas in children: Development and validation of the schema inventory for children. Journal of Behavior Therapy and Experimental Psychiatry, 41(2), 102–109. Rijkeboer, M. M., van den Bergh, H., & van den Bout, J. (2005). Stability and discriminative power of the Young Schema-Questionnaire in a Dutch clinical versus nonclinical sample. Journal of Behavior Therapy and Experimental Psychiatry, 36, 129–144. Roelofs, J., Lee, C., Ruijten, T., & Lobbestael, J. (2011). The mediating role of early maladaptive schemas in the relation between quality of attachment relationships and symptoms of depression in adolescents. Behavioural and Cognitive Psychotherapy, 39, 471–479. http://dx.doi.org/10.1017/S1352465811000117 Satorra, A., & Bentler, P. M. (2001). A scaled difference chi-square test statistic for moment structure analysis. Psychometrika, 66, 507–514. Sava, F. A. (2009). Maladaptive schemas, irrational beliefs, and their relationship with the Five-Factor Personality model. Journal of Cognitive and Behavioral Psychotherapies, 9, 135–147. Shrout, P. E., & Bolger, N. (2002). Mediation in experimental and nonexperimental studies: New procedures and recommendations. Psychological Methods, 7, 422–445. Simon, N. M., Herlands, N. N., Marks, E. H., Mancini, C., Letamendi, A., Li, Z., & Stein, M. B. (2009). Childhood maltreatment linked to greater severity and poorer quality of life and function in social anxiety disorder. Depression and Anxiety, 26, 1027–1032. http://dx.doi.org/10.1002/da.20604 Spanish Society of Epidemiology and Family and Community Medicine. (2000). Una propuesta de medida de la clase social [A proposed measure of social class]. Atención Primaria, 25, 350–363. Stallard, P. (2007). Early maladaptive schemas in children: Stability and differences between a community and a clinic referred sample. Clinical Psychology & Psychotherapy, 14, 10–18. Stallard, P., & Rayner, H. (2005). The development and preliminary evaluation of a Schema Questionnaire for Children (SQC). Behavioural and Cognitive Psychotherapy, 33, 217–224. Straus, M. A., Hamby, S. L., Finkelhor, D., Moore, D. W., & Runyan, D. (1998). Identification of child maltreatment with the Parent–Child Conflict Tactics Scales: Development and psychometric data for a national sample of American parents. Child Abuse & Neglect, 22, 249–270. Thimm, J. C. (2010a). Mediation of early maladaptive schemas between perceptions of parental rearing style and personality disorder symptoms. Journal of Behavior Therapy and Experimental Psychiatry, 41, 52–59. http://dx.doi.org/10.1016/j.jbtep.2009.10.001 Thimm, J. C. (2010b). Personality and early maladaptive schemas: A five-factor model perspective. Journal of Behavior Therapy and Experimental Psychiatry, 41, 373–380. http://dx.doi.org/10.1016/j.jbtep.2010.03.009 Wright, M. O., Crawford, E., & Del Castillo, D. (2009). Childhood emotional maltreatment and later psychological distress among college students: The mediating role of maladaptive schemas. Child Abuse & Neglect, 33, 59–68. Young, J. (1999). Cognitive therapy for personality disorders: A schema-focused approach (3rd ed.). Sarasota, FL: Professional Resources Press. Young, J. (2006). Young Schema Questionnaire-3. New York, NY: Cognitive Therapy Center. Young, J. E., Klosko, J. S., & Weishaar, M. E. (2003). Schema therapy: A practitioner’s guide. New York, NY: Guilford Press.