Journal of Adolescence 37 (2014) 1153e1160
Contents lists available at ScienceDirect
Journal of Adolescence journal homepage: www.elsevier.com/locate/jado
Bully victimization and emotional problems in adolescents: Moderation by specific cognitive coping strategies? Nadia Garnefski*, Vivian Kraaij Department of Clinical Psychology, Leiden University, The Netherlands
a b s t r a c t Keywords: Bullying Depression Anxiety Cognitive coping Coping Cognitive emotion regulation
Objective: Relationships between bully victimization and symptoms of depression/anxiety were examined. In addition, it was studied whether this relationship was moderated by specific cognitive coping strategies. Methods: Participants were 582 secondary school students who filled out online selfreport questionnaires on bully victimization, cognitive coping, and depression/anxiety. (Moderated) Multiple Regression analysis was performed. Results: Strong relationships were found between bully victimization and symptoms of depression and anxiety. On top of that, two cognitive coping strategies moderated the relationship between bullying and depression, i.e. rumination (strengthening) and positive refocusing (reducing). Cognitive coping strategies that moderated the effect of bullying on anxiety symptoms were rumination, catastrophizing (strengthening) and positive reappraisal (reducing). Conclusion: The results provide possible targets for intervention: when helping adolescents who have been bullied, maladaptive cognitive coping strategies could be assessed and challenged, while more adaptive strategies could be acquired. © 2014 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
School bullying is widely recognized as a major social problem with extensive negative consequences for the victim involved. Research has shown that child and adolescent bully victimization is associated with poorer school achievements (Nakamoto & Schwartz, 2010), lower self-esteem, more loneliness, depression and anxiety (Hawker & Boulton, 2000). Negative psychological consequences have been shown to persist into adulthood (Copeland, Wolke, Angold, & Costello, 2013). Olweus and Limber (2010, p.125) have operationalized the phenomenon of bullying as ‘aggressive behavior or intentional harm doing that is carried out repeatedly and over time in an interpersonal relationship characterized by an actual or perceived imbalance of power or strength’. Included in this operationalization are negative actions by physical contact, words, or by other ways such as gestures or intentional exclusions (Olweus & Limber, 2010). A recent, large-scale study among adolescents in 40 countries showed that 15.9% of all participating adolescents reported to have been a victim of bullying. In the same study, large differences in rates across countries were observed. Generally speaking, the highest rates of bully victimization were reported in Baltic countries; countries in north-west Europe reported lower prevalence rates than eastern European and in the majority of countries rates of victimization were higher for girls than for boys (Craig et al., 2009).
* Corresponding author. Department of Clinical Psychology, Leiden University, P.O. Box 9555, 2300 RB Leiden, The Netherlands. Tel.: þ31 715273774; fax: þ31 715274678. E-mail address: [email protected] (N. Garnefski). http://dx.doi.org/10.1016/j.adolescence.2014.07.005 0140-1971/© 2014 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
1154
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
Bullying among school children is a very old and persisting phenomenon. Despite growing (and partly successful) worldwide initiatives to reduce bullying by prevention programs (e.g. Ttofi & Farrington, 2011), it will be rather unlikely that bully behavior and bully victimization will ever be fully eradicated. From an intervention point of view it therefore remains of utmost importance to find effective strategies to help individuals cope with bully victimization in order to prevent negative psychological consequences in the short and in the long term. In a recent review study it was concluded that victims of bullying often have passive, emotionally-oriented and avoidant coping styles (Bitsch Hansen, Steenberg, Palic, & Elklit, 2012). Such passive styles, in turn, have generally also been shown to be related to depression and negative mental health and are therefore considered to be maladaptive styles (Bitsch Hansen et al., 2012). Other studies, however, concluded that even when bully victims used more problem-focused strategies, they still felt unsuccessful in solving their problems (Tenenbaum, Varjas, Meyers, & Parris, 2011). A qualitative study on the various strategies that children applied to overcome bully victimization, suggested that, regardless of the specific strategy that is used, they all appear to have in common that there is a striving for (re)experiencing dignity in the context of feeling accepted by their peers (Silberschmidt Viala, 2014). This suggests that adolescents' mental (coping) strategies may play an important role in the relationship between bully victimization and their psychological well-being. As far as we know, however, no studies have been performed that explicitly focus on the specific mental or cognitive coping strategies that adolescents use in response to bully victimization by a peer, and/or on the question whether these strategies moderate (in a positive or negative way) the relationship between bullying and depression/anxiety. In the present study we will follow that approach by focusing on the specific cognitive coping strategies that adolescents use after bully victimization. For this purpose, the theoretical framework of Garnefski, Kraaij, and Spinhoven (2001) has been chosen, in which the cognitive coping or cognitive emotion regulation strategies are defined as the conscious, mental strategies individuals use to handle the intake of emotionally arousing information (Garnefski et al., 2001; Thompson, 1991). It is assumed, that although the capability of advanced thinking and regulating emotions through thoughts and cognitions is universal, large individual differences exist in the amount of cognitive activity and in the content of thoughts of adolescents by means of which they regulate their emotions in response to life experiences, events and stressors (such as bully victimization by peers). Garnefski et al. (2001) have distinguished between nine conceptually different cognitive emotion regulation strategies that adolescents may use to regulate their emotions in response to life stress. These strategies are: 1. Self-blame, referring to thoughts of blaming yourself for what you have experienced; 2. Blaming others, referring to thoughts of putting the blame of what you have experienced on others; 3. Acceptance, referring to thoughts of accepting what you have experienced and resigning yourself to what has happened; 4. Rumination or focus on thought, referring to thinking about the feelings and thoughts associated with the negative event; 5. Catastrophizing, referring to thoughts of explicitly emphasizing the terror of an experience; 6. Refocus on planning, referring to thinking about what steps to take and how to handle the negative event. It is the cognitive part of action-focused coping, which does not automatically imply that actual behavior will follow; 7. Putting into perspective, referring to thoughts of playing down the seriousness of the event or emphasizing its relativity when compared to other events; 8. Positive reappraisal, referring to thoughts of attaching a positive meaning to the event in terms of personal growth; 9. Positive refocusing, referring to thinking about joyful and pleasant issues instead of thinking about the actual event. It has been shown that strong relationships exist between the use of these strategies and emotional problems in adolescents (e.g., Garnefski & Kraaij, 2006; Garnefski, Kraaij, & Van Etten, 2005). In general, the results suggest that by using cognitive styles such as Rumination, Catastrophizing and Self-blame people may be more vulnerable to emotional problems than others and that by using other styles, such as Positive Reappraisal or Positive Refocusing people may be more resilient. As indicated, there is a lack of research on the mentioned cognitive strategies in relation to bully victimization. However, some studies included the separate constructs as coping styles, sometimes with other names. For example Hampel, Manhal, and Hayer (2009) found that Rumination was among the maladaptive coping styles that were related to maladjustment in children and adolescents who were victimized, whereas Putting into perspective (in their study called Minimization), Positive Refocusing (Distraction) and Planning (Situation control) were among the more helpful coping styles. However, the study focused on general coping styles, and not on specific coping strategies in response to the specific context of bullying. Therefore, conclusions that can be drawn about (in)effectiveness of the use of these strategies in the context of bullying are limited. As indicated, currently no studies have been performed focusing on the moderating role that the aforementioned cognitive coping strategies may play in the relationship between bully victimization and emotional problems such as depression and anxiety in adolescents, despite the importance this information could have for identifying youngsters with increased risk of such problems and for providing targets for intervention. In a previous study that focused on younger children (9e10 years old), however, some evidence for the moderating role of certain coping strategies had been found. A strategy of problem solving - that is usually found to be beneficial for non-victimized children, appeared to exacerbate the mental health problems of the victimized children (Kochenfelder-Ladd & Skinner, 2002). Therefore, the aim of the present study is to study the relationships between bully victimization, specific cognitive coping strategies and symptoms of depression and anxiety. The first study question is: What is the direct relationship between bully victimization and severity of depression/anxiety? The first hypothesis is that a significant amount of the variance in depression/anxiety will be explained by bully victimization. The second study question is: Do specific cognitive coping strategies explain variance of depression/anxiety on top of bully victimization? The second hypothesis is, that a significant amount of the variance in depression/anxiety will be explained by specific coping strategies, on top of bully victimization.
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1155
More specifically, Rumination, Catastrophizing and Self-blame are expected to be positively related, while Positive Reappraisal and Positive Refocusing are expected to be negatively related to depression/anxiety scores. The third study question is: Is the relationship between bully victimization and severity of depression/anxiety moderated by specific cognitive coping strategies? With regard to the moderation question, we will explore the possibility that, on top of the direct effects, specific coping strategies will moderate the relationship between bully victimization and depression/anxiety. Because of the limited prior literature, no specific hypotheses were formulated with regard to this question. The questions will be studied in a general population sample of adolescents. The analyses will control for gender and negative life events. Method Sample The sample consisted of 582 13-to-16-year-old secondary school students (mean age 14 years and three months; Sd ¼ 0.84) attending a school for intermediate secondary vocational education in The Netherlands. The sample comprised 48.3% girls. Procedure In total, 69 schools for intermediate secondary vocational education were eligible for cooperation. These schools had in common that they all were on a list of schools that provided internships to students of the Amsterdam University of Applied Sciences (Hogeschool van Amsterdam, HvA). All schools were situated in the western part of The Netherlands. The schools were invited to participate by the internship coordinator of the HvA, by sending an e-mail with information on purpose and procedure of the study to the contact person. In total, 6 schools agreed to participate, with a total of twenty-seven classes. No information was available with regard to the schools that did not respond to the request for participation nor about reasons for nonparticipation. Participation consisted of administration of an online questionnaire in a classroom situation. Students filled out the computerized (self-report) questionnaire during school hours, supervised by their own teacher. They also filled out an informed consent as part of the questionnaire. There is no information on the number of students that refused participation, nor on reasons for nonparticipation. For participants younger than 16 years, parents were asked for permission to participate. Anonymity towards parents, teacher and the school was guaranteed. Ethical approval had been obtained from the ethical committee of the University. Instruments Victimization of bullying To assess bully victimization the revised Olweus Bully/Victim Questionnaire was used (Solberg & Olweus, 2003). This questionnaire consist of 40 items concerning both bully behavior and bully victimization. For the purpose of the present study only the nine items with regard to bully victimization were used. These items refer to physical, verbal, relational, and cyber bullying. Participants are asked: ‘Have you been bullied at school in the past couple of months in one or more of the following ways?’ Subsequently, the nine items are presented in the form of statements. Example statements are: ‘I was hit, kicked, pushed, shoved around, or locked indoors’ and ‘I was bullied with mean or hurtful messages, calls or pictures or in other ways on my cell phone or over the Internet (Computer)’. Answer possibilities are: 1 (‘It has not happened to me in the past couple of months’), 2 (‘Only once or twice’), 3 (‘2 or 3 times per month’), 4 (‘About once a week’), and 5 (‘Several times per week’). A total bully victimization score, reflecting the severity of bully victimization, is obtained by summing up the item scores and ranges from 9 to 45. Alpha coefficients for this questionnaire have been shown to range from 0.82 to 0.93 (Kyriakides, Kaloyirou, & Lindsay, 2006). Cognitive emotion regulation strategies To measure the specific cognitive coping strategies that adolescents used in response to maltreatment by a peer, the Cognitive Emotion Regulation Questionnaire (CERQ) was used. The CERQ is a 36-item questionnaire, consisting of the following nine conceptually and psychometrically distinct subscales: Self-blame, Other-blame, Acceptance, Rumination, Catastrophizing, Refocus on Planning, Putting into Perspective, Positive Reappraisal, Positive Refocusing (Garnefski et al., 2001, 2002). Each of the scales consists of four items and refers to what someone thinks after the experience of a threatening or stressful life event: Cognitive emotion regulation strategies are measured on a 5-point Likert scale ranging from 1 ((almost) never) to 5 ((almost) always). Individual subscale scores are obtained by summing up the scores belonging to the particular subscale or cognitive coping strategy (ranging from 4 to 20). In general, the CERQ can be used in two different ways: 1) to measure someone's cognitive coping style across different types of life events (what adolescents generally/usually think after the experience of negative or unpleasant events); and 2) to measure someone's cognitive coping strategies associated with a specific life event (what adolescents actually think in response to a particular negative event). The present study was interested in the latter category, i.e. the use of specific cognitive coping strategies in response to a specific event. In the present study, the specific event referred to ‘being maltreated
1156
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
by a peer’. To assess the cognitive strategies adolescents reported in response to ‘being maltreated by a peer’, the following instruction was provided: ‘Everyone who experiences being maltreated by a peer responds to this in his or her own way. The following questions are about what you think if someone treats you unpleasantly’. All items were stated in the present tense, referring to the current thoughts about the indicated events. Research on cognitive coping strategies, as measured by the CERQ, has shown that the subscales have good internal consistencies, with alphas ranging from .67 to .81 (Garnefski et al., 2001, 2002). Depression and anxiety Depression and anxiety were measured by two subscales of the SCL-90 (Symptom Check List: Derogatis, 1977; Dutch translation and adaptation by Arrindell & Ettema, 1986). The depression subscale consists of 15 items (item concerning loss of sexual interest was dropped, because of the age of the subjects), assessing whether and to what extent the participants report symptoms of depression; the anxiety subscale consists of 10 items, assessing whether and to what extent participants report symptoms of anxiety. Answer categories of the items range from 1 (not at all) to 5 (very much). Scale scores are obtained by summing the items belonging to the scale (Depression: range 15e75; Anxiety: range 10e50). Previous studies have reported alpha-coefficients ranging from .82 to .93 for depression and from .71 to .91 for anxiety. In addition, test-retest reliabilities are found to be good and both subscales have been found to show strong convergent validity with other conceptually related scales (Arrindell & Ettema, 1986). Life events A self-constructed checklist was used to collect data on the experience of negative life events (provided at www.cerq. leidenuniv.nl). In total, 13 life events were measured: divorce, long-lasting and/or severe physical illness of self, longlasting and/or severe physical illness of significant others, severe mental illness of significant others, death of significant others, attempted suicide of significant others, violence within the family, alcohol or drug abuse within the family, unwanted pregnancy, having been victim of crime, accident, sexual abuse, and physical abuse. A total number of negative events score was obtained (0e13). Statistical analysis First, descriptives and reliabilities of the study variables were provided. For descriptive purposes only, the bully victimization items were dichotomized by creating two groups: those who were victims of bullying 2 or 3 times per month or more (score 3,4,5) and those who were victims of bullying less than that or never (score 1,2). In the further data analyses (in the total sample), the total bully victimization score, reflecting the severity of bully victimization, was used. Subsequently, associations between the background variables (gender, age, life events) and depression/anxiety were assessed to determine whether these variables should be included as control variables in further analyses. In addition, Pearson correlations were calculated between Bully victimization and Depression/anxiety, and between Cognitive coping strategies and Anxiety/Depression. Finally to answer the questions whether specific coping strategies explain variance of depression/anxiety on top of bully victimization and whether the relationship between bully victimization and depression/anxiety is moderated by specific cognitive coping strategies, (moderated) Multiple Regression analysis was performed. Depressive symptoms/anxiety symptoms were the dependent variables. Variables were added in four steps. In the first step the control variables were entered. In the second step the bully victimization total score was entered. In the third step, the nine cognitive coping strategies were entered. In the fourth step, the interaction effects between bully victimization and cognitive coping strategies were added (Method ¼ stepwise). With regard to the interpretation of the interaction effects between bully victimization and the specific cognitive coping strategies: If a significant interaction effect was found, the sample was divided into three equal sized groups on basis of their scores on the specific cognitive strategy: low, medium, and high use of the specific strategy. Subsequently, Pearson correlations were calculated for each of the three groups and compared to each other. Results In Table 1 bully victimization characteristics of the participants are presented. In the first column the items of the Olweus Bully questionnaire are shown, reflecting the different forms of bullying; in the second column the mean scores of the items are presented; in the third column the percentages per item are given of the adolescents who were bullied two or three times per month or more (for this purpose the items were dichotomized). The type of bullying reported most often was bullied by being called mean names, followed by bullied by lies/false rumors being told. Significant gender differences were only found with regard to telling lies or false rumors: 12.9% of the girls versus 6.7% of the boys reported that they had experienced this type of bullying two or three times per month or more (c2(1) ¼ 6.36; p ¼ .016; not in table). In total, 137 participants reported to have experienced one or more of the specific types of bullying at least two or three times per month. The mean total bully victimization score of the whole sample was 11.47 (Sd ¼ 4.05), while the range of actual scores was from 9 to 45. The alpha reliability of the 9 bully items was .80. Table 2 presents the descriptives of the other study variables. Alpha reliabilities of the depression and anxiety scales were .93 and .90, respectively. The alpha reliabilities from the CERQ scales ranged from .72 to .82.
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1157
Table 1 Bully victimization characteristics of participants (N ¼ 582). Bully victimization
Bullied by being called mean names Bullied by exclusion Bullied by hitting, kicking etc Bullied by lies/false rumors being told Bullied by taking things away Bullied by threatening/forcing Bullied by comments about race/color Bullied by sexual comments Bullied via cell phone or internet Number/percentage that was bullied two or three times per month or more by means of one or more of the above Mean and Standard deviation of Bully victimization total score (9e45)
Mean and standard deviation per item (score range 1-5)
Number/percentage that was bullied two or three times per month or more (adolescents with score 3,4, or 5)
M (Sd)
N (%)
1.71 1.30 1.25 1.44 1.13 1.07 1.23 1.20 1.14
(1.08) (0.83) (0.75) (0.88) (0.48) (0.34) (0.68) (0.65) (0.52)
90 35 33 56 11 4 32 25 13 137
(15.3%) (6.1%) (5.7%) (9.6%) (1.9%) (0.7%) (5.5%) (4.3%) (2.2%) (23.5%)
11.47 (4.05)
Next, the bivariate relationships between the background variables (gender, age, life events) and symptoms of depression and anxiety were tested (no table). With regard to gender: T-tests pointed out that girls reported significantly more symptoms of depression and anxiety than boys (depression: t(578) ¼ 5.36; p < .001; anxiety: t(578) ¼ 4.25; p < .001). With regard to age: No significant relationships were found between age and symptoms of depression/anxiety (respectively r ¼ .01 for depression and r ¼ .03 for anxiety). With regard to life events: Significant positive Pearson correlations were found between ‘number of life events’ and symptoms of depression and anxiety (respectively r ¼ .49 with depression and r ¼ .50 with anxiety). Therefore, only gender and ‘number of life events’ were included as control variables in the Multiple Regression analysis (MRA). In addition, Table 3 presents the bivariate Pearson correlations among bully victimization total score, depression and anxiety symptoms and cognitive coping strategies. Correlations between bully victimization total score and depression/ anxiety were .59 and .57, respectively. Bivariate correlations between cognitive coping strategies and depression ranged from .03 (Positive refocusing) to .61 (Rumination). The same pattern was observed for the correlations between cognitive coping strategies and anxiety symptoms, ranging from .01 (Positive Refocusing) to .55. The highest Pearson correlations (between .45 and .61) were found between cognitive coping strategies rumination, catastrophizing, and self-blame and both types of emotional problems: depression and anxiety. Other significant, but lower correlations with depression and anxiety were found for other-blame and acceptance (between .31 and .35). Although the correlations between planning and depression/ anxiety also were significant (.16 and .14, respectively), they only explained a marginal amount of the variance. Also the Pearson correlations among cognitive coping strategies can be found in Table 3. Correlations ranged from .16 (catastrophizing and positive refocusing) to .68 (self-blame and rumination). Although the cognitive coping strategies were significantly related, there was no evidence of logical or statistical problems due to multicollinearity. In general, correlation values of .90 and higher are considered indicative of multicollinearity of the data (Tabachnick & Fidell, 2001). Subsequently, MRA was performed with depressive symptoms as outcome variable (Table 4). First, the background variables gender and ‘number of life events’ were entered into the regression analysis as the first block (method ¼ enter), in order to control for these variables. Subsequently the variable bully victimization total score was entered. The nine cognitive strategies were entered as the third block (method ¼ enter). Finally, the interaction effects between bully victimization and cognitive strategies were added (method ¼ stepwise). The total model explained 63% of the variance of depressive symptoms (p < .001). After partialling out the effects of gender, life events and cognitive strategies, bully victimization remained a significant direct predictor of depressive symptoms (p < .001). With regard to the cognitive strategies, significant (direct) positive
Table 2 Descriptives and alpha reliabilities of depression, anxiety and cognitive coping scales (N ¼ 582).
Depressive symptoms Anxiety symptoms Cognitive coping Selfblame Otherblame Acceptance Rumination Catastrophizing Planning Putting into Perspective Positive Reappraisal Positive Refocusing
M (Sd)
Range
Cronbach's a
21.12 (9.64) 13.67 (5.61)
14e70 10e50
.93 .90
4e19 4e20 4e20 4e20 4e20 4e20 4e20 4e20 4e20
.74 .72 .72 .82 .74 .82 .77 .72 .81
7.51 6.29 8.73 7.89 6.33 9.68 9.72 9.23 9.83
(3.01) (2.47) (3.39) (3.52) (2.88) (3.78) (3.90) (3.57) (4.10)
1158
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
Table 3 Correlations among study variables (N ¼ 582).
1 Bully victimization 2 Depression 3 Anxiety Cognitive coping 4 Selfblame 5 Otherblame 6 Acceptance 7 Rumination 8 Catastrophizing 9 Planning 10 Putting into Perspective 11 Positive Reappraisal 12 Positive Refocusing
1 r
2 r
3 r
e .59*** .57***
e
.36*** .26*** .24*** .42*** .41*** .18*** .04 .11** .03
..49*** .32*** .35*** .61*** .53*** .16*** .05 .04 .03
.87***
4 r
5 r
6 r
7 r
8 r
9 r
10 r
11 r
12 r
e .36*** .64*** .68*** .53*** .51*** .38*** .41*** .24***
e .38*** .47*** .50*** .46*** .31*** .35*** .26***
e .62*** .49*** .58*** .53*** .56*** .45***
e .63*** .52*** .28*** .36*** .21***
e .35*** .23*** .30*** .16***
e .56*** .70*** .54***
.67*** ..57***
e .60***
e
e .45*** .33*** .31*** .55*** .52*** .14** .04 .04 .01
***: p < .001; **: p < .01; *: p < .05.
‘predictors’ of depressive symptoms were self-blame, rumination and catastrophizing (higher use of these strategies related to more depressive symptoms). Significant (direct) negative ‘predictors’ were planning and positive reappraisal (see Table 4). The first significant interaction effect that was found was: being bullied by rumination. To understand this effect the sample was divided in three equal sized groups on basis of their rumination scores (low, medium, high). In the low rumination group the Pearson correlation between bully victimization and depressive symptoms (corrected for gender and life events) was .20. In the medium rumination group the Pearson correlation was .35 and in the high rumination group the correlation was .55, showing that higher rumination strengthened the association between bully victimization and depressive symptoms. Also a significant (negative) interaction effect was found for being bullied by positive refocusing. Again, the sample was divided in three equal sized groups on basis of their positive refocusing scores. In the low positive refocusing group the correlation (corrected for gender and life events) was .60, in the medium positive refocusing group the correlation was .44 and in the high positive refusing group the correlation was .38. This indicated that higher positive refocusing lowered the association between bully victimization and depressive symptoms. Next, MRA was performed with anxiety symptoms as dependent variable (Table 4). After controlling for gender and life events in the first block, and entering the bully victimization total score in the second block, the nine cognitive strategies were entered. Direct, positive significant effects were found for self-blame, other-blame, rumination, and catastrophizing, whereas negative significant effects were found for planning and positive reappraisal. In the next block, the interaction effects were added (stepwise). The total model explained 58% of the variance of depressive symptoms (p < .001). Significant interaction were found for bully victimization with rumination, catastrophizing and positive reappraisal.
Table 4 MRA on Depression and Anxiety, with interaction terms: method enter. Depression
b Background variables Gender Life events Bully victimization Cognitive coping Selfblame Otherblame Acceptance Rumination Catastrophizing Planning Putting into Perspective Positive Reappraisal Positive Refocusing Significant moderators Bullied by rumination Bullied by Positive Refocusing Bullied by Catastrophizing Bullied by Positive Reappraisal Model Explained variance (R2) ***: p < .001; **: p < .01; *:p < .05.
Anxiety
b
t
t
.09 .21 .22
3.41** 7.24*** 6.19***
.07 .24 .21
2.34* 7.70*** 5.02***
.14 .06 .02 .28 .09 .09 .01 .17 .03
3.52*** 1.87 0.47 6.24*** 2.52* 2.12* 0.34 4.00*** 0.85
.14 .10 .02 .22 .08 .10 .01 .16 .02
3.40** 2.75** 0.41 4.42*** 1.99* 2.27* 0.30 3.40** 0.65
.13 .08
3.99** 2.95**
F(14,526) ¼ 67.94; p ¼ .000 R2 ¼ .63
.09
2.25*
.14 3.34** .10 3.04** F(15,561) ¼ 51.24; p ¼ .000 R2 ¼ .58
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1159
With regard to the interpretation of the interaction effect of bully victimization and rumination: the sample was divided in three equal sized groups on basis of their rumination scores (low, medium, high). The following correlations were found in the low, medium and high rumination groups (after correction for life events and gender): .24, .37 and .44, showing that higher rumination also strengthened the association between bully victimization and anxiety symptoms. With regard to the interaction effect of bully victimization and catastrophizing, again the sample was divided in three equal sized groups (on basis of catastrophizing scores). The sample correlations were .21, .47, and .46 in the low, medium and high score groups, respectively, showing that both medium and high catastrophizing strengthened the association between bully victimization and anxiety symptoms. With regard to positive reappraisal, three equal sized groups were also created. Correlations between bully victimization and anxiety scores were .57 in the low group, .52 in the medium group, and .42 in the high score group. These results indicated that higher positive reappraisal lowered the association between bully victimization and anxiety. Discussion The present study focused on the role that specific cognitive coping strategies played in the relationship between bully victimization and symptoms of depression and anxiety. First, a strong direct relationship was found between bully victimization and both depression and anxiety symptoms, confirming results of previous studies (Hawker & Boulton, 2000). On top of this direct effect of bullying, specific cognitive coping strategies added a significant amount to the explained variance of depression and anxiety. The strongest direct positive effects were found for self-blame and rumination, whereas the strongest negative effects was found for positive reappraisal. No significant effect was found for Catastrophizing in the regression analysis, however a strong Pearson correlation was found between Catastrophizing and depression/anxiety. These results confirmed the findings of previous studies suggesting that by using cognitive styles such as rumination, catastrophizing and self-blame people may be more vulnerable to emotional problems than others, while by using other styles, such as positive reappraisal people may be more resilient (eg., Garnefski et al., 2001). As far as we know, no studies had been performed before that focused on the moderating role that specific cognitive coping strategies might play in the relationship between bully victimization and symptoms of depression. The present study filled in this gap by identifying two moderators in the relationship between bullying and depression. The first moderator was rumination. Using this strategy appeared to strengthen the effect of bullying on depression. The second moderator was positive refocusing. This strategy appeared to reduce the effect of bullying on depression. Coping strategies that moderated the effect of bullying on anxiety symptoms were rumination, catastrophizing (strengthening the effect on anxiety) and positive reappraisal (reducing the effect on anxiety). Before discussing the possible implications of the results, it is important to address some limitations. One important issue was that the detection of psychological distress as well as the assessment of bully victimization and cognitive strategies was made on basis of self-report. In order to be able to draw firm conclusions about cognitive coping strategies contributing to risk and resilience of emotional problems after bully victimization, further studies are necessary. In such studies more qualitative methods should be applied, for example by using in depth interviews and including narrative measures or daily diaries. In addition, all students of our sample were attending schools for intermediate vocational education. Another issue is that the results of the present study are based on cross-sectional data. Therefore, it is important to acknowledge that no conclusions can be drawn about causality or directions of influence. For purposes of generalizability, the study should be repeated with other samples and with other methods, while prospective elements should be included. If the results of our study can be confirmed in further studies, they might have important practical implications. Firstly, assessment of maladaptive cognitive coping strategies might help to identify adolescents with increased risk of developing symptoms of depression and/or anxiety. Secondly, and even more importantly, the results might provide targets for intervention. One of the important conclusions that was already drawn in the review study of Greenberg, Domitrovich and Bumbarger (2001) concerning the prevention of mental disorders in school-aged children, was that (preventive) interventions should preferably be focused at risk and protective factors rather than to categorical problem behaviors. During the past decades (including the present study) bully victimization has been clearly found to place an adolescent at increased risk for psychopathology (Hawker & Boulton, 2000). The present study could add to this conclusion that on top of that certain maladaptive cognitive strategies might increase the risk for psychopathology in adolescents, whereas other cognitive strategies might be able to reduce the risk. An important target for interventions might therefore be to prevent maladaptive cognitive coping strategies from turning into long-established and difficult-to-change styles by reducing non-adaptive strategies and acquiring more adaptive strategies. The results of the present study might give specific suggestions for the content of such interventions. Firstly, when helping an adolescent to cope with bully victimization, it could be important for health care professionals to assess to what extent the adolescent uses self-blaming, rumination, or catastrophizing in response to bully victimization. Secondly, the adolescent could be assisted to become more aware of and to challenge these maladaptive cognitive coping patterns and to acquire new, more adaptive strategies such as positive refocusing or positive reappraisal. Although specific interventions on changing cognitive coping strategies are not available yet, these could easily be integrated in existing Cognitive Behavioral Therapies, which are widely used and evidence based (Hofmann, Asnaani, Vonk, Sawyer, & Fang, 2012). However, as mentioned before, before drawing firm conclusions, further studies in other samples and with other research methods that focus on the same research questions, are necessary. If the results are confirmed, they carry important
1160
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
implications for the focus and content of intervention and prevention of mental health problems after the experience of bullying by adolescents. Acknowledgments The authors would like to thank the students who participated in this research project: David Kooij, Rob Krens, and Birgit Olsen. References Arrindell, W. A., & Ettema, J. H. M. (1986). SCL-90, Handleiding bij een multidimensionele psychopathologie-indicator [SCL-90, Manual for a multidimensional psychopathology indicator]. Lisse: Swets & Zeitlinger B.V. Bitsch Hansen, T., Steenberg, L. M., Palic, S., & Elklit, A. (2012). A review of psychological factors related to bullying victimization in schools. Aggression and Violent Behavior, 17, 383e387. Copeland, W. E., Wolke, D., Angold, A., & Costello, J. (2013). Adult Psychiatric outcomes of bullying and being bullied by peers in Childhood and Adolescence. JAMA Psychiatry, 70, 419e426. Craig, W., Harel-Fisch, Y., Fogel-Grinvald, H., Dostaler, S., Hetland, J., Simons-Morton, B., &, HBSC Bullying writing group. (2009). A cross-national profile of bullying and victimization among adolescents in 40 countries. International Journal of Public Health, S216eS224. Derogatis, L. R. (1977). SCL-90: Administration, scoring and procedures manual-I for the r(evised) version. Baltimore: John Hopkins University School of Medicine, Clinical Psychometrics Research Unit. Garnefski, N., & Kraaij, V. (2006). Relationships between cognitive emotion regulation strategies and depressive symptoms: a comparative study of five specific samples. Personality and Individual Differences, 40, 1659e1669. Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative life events, cognitive emotion regulation and emotional problems. Personality and Individual Differences, 30, 1311e1327. Garnefski, N., Kraaij, V., & Spinhoven, P. (2002). CERQ: Manual for the use of the cognitive Emotion Regulation Questionnaire: A questionnaire measuring cognitive coping strategies. Leiderdorp, The Netherlands: DATEC. Garnefski, N., Kraaij, V., & Van Etten, M. (2005). Specificity of relations between adolescents’ cognitive emotion regulation strategies and internalizing and externalizing psychopathology. Journal of Adolescence, 28, 619e631. Greenberg, M. T., Domitrovic, C., & Bumbarger, B. (2001). The prevention of mental disorders in school-aged children: current state of the field. Prevention & Treatment, 4. article 1. Hampel, P., Manhal, S., & Hayer, T. (2009). Direct and relational bullying among children and adolescents: coping and psychological adjustment. School Psychology International, 30, 474e490. Hawker, D. S. J., & Boulton, M. (2000). Twenty years' research on peer victimization and psychosocial maladjustment: a meta-analytic review of crosssectional studies. Journal of Child Psychology and Psychiatry and Allied Disciplines, 41, 441e455. Hofmann, S. G., Asnaani, A., Vonk, I. J. J., Sawyer, A. T., & Fang, A. (2012). The efficacy of cognitive behavioral therapy: a review of meta-analyses. Cognitive Therapy Research, 36, 427e440. Kochenfelder-Ladd, B., & Skinner, K. (2002). Children's coping strategies: moderators of the effects of peer victimization? Developmental Psychology, 38, 267e278. Kyriakides, L., Kaloyirou, C., & Lindsay, G. (2006). An analysis of the Revised Olweus Bully/Victim Questionnaire using the Rasch measurement model. British Journal of Educational Psychology, 76, 781e801. Nakamoto, J., & Schwartz, D. (2010). Is peer victimization associated with academic achievement? A meta-analytic review. Social Development, 19, 221e242. Olweus, D., & Limber, S. P. (2010). Bullying in school: evaluation and dissemination of the Olweus Bullying Prevention Program. American Journal of Orthopsychiatry, 80, 124e134. Silberschmidt Viala, E. (2014). The fighter, the punk, and the clown: how to overcome the position of victim of bullying. Advance online publication Childhood. http://dx.doi.org/10.1177/0907568214521845. Solberg, M. E., & Olweus, D. (2003). Prevalence estimation of school bullying with the Olweus Bully/Victim Questionnaire. Aggressive Behavior, 29, 239e268. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham Heights, USA: Allyn & Bacon. Tenenbaum, L. S., Varjas, K., Meyers, J., & Parris, L. (2011). Coping strategies and perceived effectiveness in fourth through eighth grade victims of bullying. School Psychology International, 32, 263e286. Thompson, R. A. (1991). Emotional regulation and emotional development. Educational Psychology Review, 3, 269e307. Ttofi, M. M., & Farrington, D. P. (2011). Effectiveness of school-based programs to reduce bullying: a systematic and meta-analytic review. Journal of Experimental Criminology, 7, 27e56.
Contents lists available at ScienceDirect
Journal of Adolescence journal homepage: www.elsevier.com/locate/jado
Bully victimization and emotional problems in adolescents: Moderation by specific cognitive coping strategies? Nadia Garnefski*, Vivian Kraaij Department of Clinical Psychology, Leiden University, The Netherlands
a b s t r a c t Keywords: Bullying Depression Anxiety Cognitive coping Coping Cognitive emotion regulation
Objective: Relationships between bully victimization and symptoms of depression/anxiety were examined. In addition, it was studied whether this relationship was moderated by specific cognitive coping strategies. Methods: Participants were 582 secondary school students who filled out online selfreport questionnaires on bully victimization, cognitive coping, and depression/anxiety. (Moderated) Multiple Regression analysis was performed. Results: Strong relationships were found between bully victimization and symptoms of depression and anxiety. On top of that, two cognitive coping strategies moderated the relationship between bullying and depression, i.e. rumination (strengthening) and positive refocusing (reducing). Cognitive coping strategies that moderated the effect of bullying on anxiety symptoms were rumination, catastrophizing (strengthening) and positive reappraisal (reducing). Conclusion: The results provide possible targets for intervention: when helping adolescents who have been bullied, maladaptive cognitive coping strategies could be assessed and challenged, while more adaptive strategies could be acquired. © 2014 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
School bullying is widely recognized as a major social problem with extensive negative consequences for the victim involved. Research has shown that child and adolescent bully victimization is associated with poorer school achievements (Nakamoto & Schwartz, 2010), lower self-esteem, more loneliness, depression and anxiety (Hawker & Boulton, 2000). Negative psychological consequences have been shown to persist into adulthood (Copeland, Wolke, Angold, & Costello, 2013). Olweus and Limber (2010, p.125) have operationalized the phenomenon of bullying as ‘aggressive behavior or intentional harm doing that is carried out repeatedly and over time in an interpersonal relationship characterized by an actual or perceived imbalance of power or strength’. Included in this operationalization are negative actions by physical contact, words, or by other ways such as gestures or intentional exclusions (Olweus & Limber, 2010). A recent, large-scale study among adolescents in 40 countries showed that 15.9% of all participating adolescents reported to have been a victim of bullying. In the same study, large differences in rates across countries were observed. Generally speaking, the highest rates of bully victimization were reported in Baltic countries; countries in north-west Europe reported lower prevalence rates than eastern European and in the majority of countries rates of victimization were higher for girls than for boys (Craig et al., 2009).
* Corresponding author. Department of Clinical Psychology, Leiden University, P.O. Box 9555, 2300 RB Leiden, The Netherlands. Tel.: þ31 715273774; fax: þ31 715274678. E-mail address: [email protected] (N. Garnefski). http://dx.doi.org/10.1016/j.adolescence.2014.07.005 0140-1971/© 2014 The Foundation for Professionals in Services for Adolescents. Published by Elsevier Ltd. All rights reserved.
1154
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
Bullying among school children is a very old and persisting phenomenon. Despite growing (and partly successful) worldwide initiatives to reduce bullying by prevention programs (e.g. Ttofi & Farrington, 2011), it will be rather unlikely that bully behavior and bully victimization will ever be fully eradicated. From an intervention point of view it therefore remains of utmost importance to find effective strategies to help individuals cope with bully victimization in order to prevent negative psychological consequences in the short and in the long term. In a recent review study it was concluded that victims of bullying often have passive, emotionally-oriented and avoidant coping styles (Bitsch Hansen, Steenberg, Palic, & Elklit, 2012). Such passive styles, in turn, have generally also been shown to be related to depression and negative mental health and are therefore considered to be maladaptive styles (Bitsch Hansen et al., 2012). Other studies, however, concluded that even when bully victims used more problem-focused strategies, they still felt unsuccessful in solving their problems (Tenenbaum, Varjas, Meyers, & Parris, 2011). A qualitative study on the various strategies that children applied to overcome bully victimization, suggested that, regardless of the specific strategy that is used, they all appear to have in common that there is a striving for (re)experiencing dignity in the context of feeling accepted by their peers (Silberschmidt Viala, 2014). This suggests that adolescents' mental (coping) strategies may play an important role in the relationship between bully victimization and their psychological well-being. As far as we know, however, no studies have been performed that explicitly focus on the specific mental or cognitive coping strategies that adolescents use in response to bully victimization by a peer, and/or on the question whether these strategies moderate (in a positive or negative way) the relationship between bullying and depression/anxiety. In the present study we will follow that approach by focusing on the specific cognitive coping strategies that adolescents use after bully victimization. For this purpose, the theoretical framework of Garnefski, Kraaij, and Spinhoven (2001) has been chosen, in which the cognitive coping or cognitive emotion regulation strategies are defined as the conscious, mental strategies individuals use to handle the intake of emotionally arousing information (Garnefski et al., 2001; Thompson, 1991). It is assumed, that although the capability of advanced thinking and regulating emotions through thoughts and cognitions is universal, large individual differences exist in the amount of cognitive activity and in the content of thoughts of adolescents by means of which they regulate their emotions in response to life experiences, events and stressors (such as bully victimization by peers). Garnefski et al. (2001) have distinguished between nine conceptually different cognitive emotion regulation strategies that adolescents may use to regulate their emotions in response to life stress. These strategies are: 1. Self-blame, referring to thoughts of blaming yourself for what you have experienced; 2. Blaming others, referring to thoughts of putting the blame of what you have experienced on others; 3. Acceptance, referring to thoughts of accepting what you have experienced and resigning yourself to what has happened; 4. Rumination or focus on thought, referring to thinking about the feelings and thoughts associated with the negative event; 5. Catastrophizing, referring to thoughts of explicitly emphasizing the terror of an experience; 6. Refocus on planning, referring to thinking about what steps to take and how to handle the negative event. It is the cognitive part of action-focused coping, which does not automatically imply that actual behavior will follow; 7. Putting into perspective, referring to thoughts of playing down the seriousness of the event or emphasizing its relativity when compared to other events; 8. Positive reappraisal, referring to thoughts of attaching a positive meaning to the event in terms of personal growth; 9. Positive refocusing, referring to thinking about joyful and pleasant issues instead of thinking about the actual event. It has been shown that strong relationships exist between the use of these strategies and emotional problems in adolescents (e.g., Garnefski & Kraaij, 2006; Garnefski, Kraaij, & Van Etten, 2005). In general, the results suggest that by using cognitive styles such as Rumination, Catastrophizing and Self-blame people may be more vulnerable to emotional problems than others and that by using other styles, such as Positive Reappraisal or Positive Refocusing people may be more resilient. As indicated, there is a lack of research on the mentioned cognitive strategies in relation to bully victimization. However, some studies included the separate constructs as coping styles, sometimes with other names. For example Hampel, Manhal, and Hayer (2009) found that Rumination was among the maladaptive coping styles that were related to maladjustment in children and adolescents who were victimized, whereas Putting into perspective (in their study called Minimization), Positive Refocusing (Distraction) and Planning (Situation control) were among the more helpful coping styles. However, the study focused on general coping styles, and not on specific coping strategies in response to the specific context of bullying. Therefore, conclusions that can be drawn about (in)effectiveness of the use of these strategies in the context of bullying are limited. As indicated, currently no studies have been performed focusing on the moderating role that the aforementioned cognitive coping strategies may play in the relationship between bully victimization and emotional problems such as depression and anxiety in adolescents, despite the importance this information could have for identifying youngsters with increased risk of such problems and for providing targets for intervention. In a previous study that focused on younger children (9e10 years old), however, some evidence for the moderating role of certain coping strategies had been found. A strategy of problem solving - that is usually found to be beneficial for non-victimized children, appeared to exacerbate the mental health problems of the victimized children (Kochenfelder-Ladd & Skinner, 2002). Therefore, the aim of the present study is to study the relationships between bully victimization, specific cognitive coping strategies and symptoms of depression and anxiety. The first study question is: What is the direct relationship between bully victimization and severity of depression/anxiety? The first hypothesis is that a significant amount of the variance in depression/anxiety will be explained by bully victimization. The second study question is: Do specific cognitive coping strategies explain variance of depression/anxiety on top of bully victimization? The second hypothesis is, that a significant amount of the variance in depression/anxiety will be explained by specific coping strategies, on top of bully victimization.
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1155
More specifically, Rumination, Catastrophizing and Self-blame are expected to be positively related, while Positive Reappraisal and Positive Refocusing are expected to be negatively related to depression/anxiety scores. The third study question is: Is the relationship between bully victimization and severity of depression/anxiety moderated by specific cognitive coping strategies? With regard to the moderation question, we will explore the possibility that, on top of the direct effects, specific coping strategies will moderate the relationship between bully victimization and depression/anxiety. Because of the limited prior literature, no specific hypotheses were formulated with regard to this question. The questions will be studied in a general population sample of adolescents. The analyses will control for gender and negative life events. Method Sample The sample consisted of 582 13-to-16-year-old secondary school students (mean age 14 years and three months; Sd ¼ 0.84) attending a school for intermediate secondary vocational education in The Netherlands. The sample comprised 48.3% girls. Procedure In total, 69 schools for intermediate secondary vocational education were eligible for cooperation. These schools had in common that they all were on a list of schools that provided internships to students of the Amsterdam University of Applied Sciences (Hogeschool van Amsterdam, HvA). All schools were situated in the western part of The Netherlands. The schools were invited to participate by the internship coordinator of the HvA, by sending an e-mail with information on purpose and procedure of the study to the contact person. In total, 6 schools agreed to participate, with a total of twenty-seven classes. No information was available with regard to the schools that did not respond to the request for participation nor about reasons for nonparticipation. Participation consisted of administration of an online questionnaire in a classroom situation. Students filled out the computerized (self-report) questionnaire during school hours, supervised by their own teacher. They also filled out an informed consent as part of the questionnaire. There is no information on the number of students that refused participation, nor on reasons for nonparticipation. For participants younger than 16 years, parents were asked for permission to participate. Anonymity towards parents, teacher and the school was guaranteed. Ethical approval had been obtained from the ethical committee of the University. Instruments Victimization of bullying To assess bully victimization the revised Olweus Bully/Victim Questionnaire was used (Solberg & Olweus, 2003). This questionnaire consist of 40 items concerning both bully behavior and bully victimization. For the purpose of the present study only the nine items with regard to bully victimization were used. These items refer to physical, verbal, relational, and cyber bullying. Participants are asked: ‘Have you been bullied at school in the past couple of months in one or more of the following ways?’ Subsequently, the nine items are presented in the form of statements. Example statements are: ‘I was hit, kicked, pushed, shoved around, or locked indoors’ and ‘I was bullied with mean or hurtful messages, calls or pictures or in other ways on my cell phone or over the Internet (Computer)’. Answer possibilities are: 1 (‘It has not happened to me in the past couple of months’), 2 (‘Only once or twice’), 3 (‘2 or 3 times per month’), 4 (‘About once a week’), and 5 (‘Several times per week’). A total bully victimization score, reflecting the severity of bully victimization, is obtained by summing up the item scores and ranges from 9 to 45. Alpha coefficients for this questionnaire have been shown to range from 0.82 to 0.93 (Kyriakides, Kaloyirou, & Lindsay, 2006). Cognitive emotion regulation strategies To measure the specific cognitive coping strategies that adolescents used in response to maltreatment by a peer, the Cognitive Emotion Regulation Questionnaire (CERQ) was used. The CERQ is a 36-item questionnaire, consisting of the following nine conceptually and psychometrically distinct subscales: Self-blame, Other-blame, Acceptance, Rumination, Catastrophizing, Refocus on Planning, Putting into Perspective, Positive Reappraisal, Positive Refocusing (Garnefski et al., 2001, 2002). Each of the scales consists of four items and refers to what someone thinks after the experience of a threatening or stressful life event: Cognitive emotion regulation strategies are measured on a 5-point Likert scale ranging from 1 ((almost) never) to 5 ((almost) always). Individual subscale scores are obtained by summing up the scores belonging to the particular subscale or cognitive coping strategy (ranging from 4 to 20). In general, the CERQ can be used in two different ways: 1) to measure someone's cognitive coping style across different types of life events (what adolescents generally/usually think after the experience of negative or unpleasant events); and 2) to measure someone's cognitive coping strategies associated with a specific life event (what adolescents actually think in response to a particular negative event). The present study was interested in the latter category, i.e. the use of specific cognitive coping strategies in response to a specific event. In the present study, the specific event referred to ‘being maltreated
1156
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
by a peer’. To assess the cognitive strategies adolescents reported in response to ‘being maltreated by a peer’, the following instruction was provided: ‘Everyone who experiences being maltreated by a peer responds to this in his or her own way. The following questions are about what you think if someone treats you unpleasantly’. All items were stated in the present tense, referring to the current thoughts about the indicated events. Research on cognitive coping strategies, as measured by the CERQ, has shown that the subscales have good internal consistencies, with alphas ranging from .67 to .81 (Garnefski et al., 2001, 2002). Depression and anxiety Depression and anxiety were measured by two subscales of the SCL-90 (Symptom Check List: Derogatis, 1977; Dutch translation and adaptation by Arrindell & Ettema, 1986). The depression subscale consists of 15 items (item concerning loss of sexual interest was dropped, because of the age of the subjects), assessing whether and to what extent the participants report symptoms of depression; the anxiety subscale consists of 10 items, assessing whether and to what extent participants report symptoms of anxiety. Answer categories of the items range from 1 (not at all) to 5 (very much). Scale scores are obtained by summing the items belonging to the scale (Depression: range 15e75; Anxiety: range 10e50). Previous studies have reported alpha-coefficients ranging from .82 to .93 for depression and from .71 to .91 for anxiety. In addition, test-retest reliabilities are found to be good and both subscales have been found to show strong convergent validity with other conceptually related scales (Arrindell & Ettema, 1986). Life events A self-constructed checklist was used to collect data on the experience of negative life events (provided at www.cerq. leidenuniv.nl). In total, 13 life events were measured: divorce, long-lasting and/or severe physical illness of self, longlasting and/or severe physical illness of significant others, severe mental illness of significant others, death of significant others, attempted suicide of significant others, violence within the family, alcohol or drug abuse within the family, unwanted pregnancy, having been victim of crime, accident, sexual abuse, and physical abuse. A total number of negative events score was obtained (0e13). Statistical analysis First, descriptives and reliabilities of the study variables were provided. For descriptive purposes only, the bully victimization items were dichotomized by creating two groups: those who were victims of bullying 2 or 3 times per month or more (score 3,4,5) and those who were victims of bullying less than that or never (score 1,2). In the further data analyses (in the total sample), the total bully victimization score, reflecting the severity of bully victimization, was used. Subsequently, associations between the background variables (gender, age, life events) and depression/anxiety were assessed to determine whether these variables should be included as control variables in further analyses. In addition, Pearson correlations were calculated between Bully victimization and Depression/anxiety, and between Cognitive coping strategies and Anxiety/Depression. Finally to answer the questions whether specific coping strategies explain variance of depression/anxiety on top of bully victimization and whether the relationship between bully victimization and depression/anxiety is moderated by specific cognitive coping strategies, (moderated) Multiple Regression analysis was performed. Depressive symptoms/anxiety symptoms were the dependent variables. Variables were added in four steps. In the first step the control variables were entered. In the second step the bully victimization total score was entered. In the third step, the nine cognitive coping strategies were entered. In the fourth step, the interaction effects between bully victimization and cognitive coping strategies were added (Method ¼ stepwise). With regard to the interpretation of the interaction effects between bully victimization and the specific cognitive coping strategies: If a significant interaction effect was found, the sample was divided into three equal sized groups on basis of their scores on the specific cognitive strategy: low, medium, and high use of the specific strategy. Subsequently, Pearson correlations were calculated for each of the three groups and compared to each other. Results In Table 1 bully victimization characteristics of the participants are presented. In the first column the items of the Olweus Bully questionnaire are shown, reflecting the different forms of bullying; in the second column the mean scores of the items are presented; in the third column the percentages per item are given of the adolescents who were bullied two or three times per month or more (for this purpose the items were dichotomized). The type of bullying reported most often was bullied by being called mean names, followed by bullied by lies/false rumors being told. Significant gender differences were only found with regard to telling lies or false rumors: 12.9% of the girls versus 6.7% of the boys reported that they had experienced this type of bullying two or three times per month or more (c2(1) ¼ 6.36; p ¼ .016; not in table). In total, 137 participants reported to have experienced one or more of the specific types of bullying at least two or three times per month. The mean total bully victimization score of the whole sample was 11.47 (Sd ¼ 4.05), while the range of actual scores was from 9 to 45. The alpha reliability of the 9 bully items was .80. Table 2 presents the descriptives of the other study variables. Alpha reliabilities of the depression and anxiety scales were .93 and .90, respectively. The alpha reliabilities from the CERQ scales ranged from .72 to .82.
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1157
Table 1 Bully victimization characteristics of participants (N ¼ 582). Bully victimization
Bullied by being called mean names Bullied by exclusion Bullied by hitting, kicking etc Bullied by lies/false rumors being told Bullied by taking things away Bullied by threatening/forcing Bullied by comments about race/color Bullied by sexual comments Bullied via cell phone or internet Number/percentage that was bullied two or three times per month or more by means of one or more of the above Mean and Standard deviation of Bully victimization total score (9e45)
Mean and standard deviation per item (score range 1-5)
Number/percentage that was bullied two or three times per month or more (adolescents with score 3,4, or 5)
M (Sd)
N (%)
1.71 1.30 1.25 1.44 1.13 1.07 1.23 1.20 1.14
(1.08) (0.83) (0.75) (0.88) (0.48) (0.34) (0.68) (0.65) (0.52)
90 35 33 56 11 4 32 25 13 137
(15.3%) (6.1%) (5.7%) (9.6%) (1.9%) (0.7%) (5.5%) (4.3%) (2.2%) (23.5%)
11.47 (4.05)
Next, the bivariate relationships between the background variables (gender, age, life events) and symptoms of depression and anxiety were tested (no table). With regard to gender: T-tests pointed out that girls reported significantly more symptoms of depression and anxiety than boys (depression: t(578) ¼ 5.36; p < .001; anxiety: t(578) ¼ 4.25; p < .001). With regard to age: No significant relationships were found between age and symptoms of depression/anxiety (respectively r ¼ .01 for depression and r ¼ .03 for anxiety). With regard to life events: Significant positive Pearson correlations were found between ‘number of life events’ and symptoms of depression and anxiety (respectively r ¼ .49 with depression and r ¼ .50 with anxiety). Therefore, only gender and ‘number of life events’ were included as control variables in the Multiple Regression analysis (MRA). In addition, Table 3 presents the bivariate Pearson correlations among bully victimization total score, depression and anxiety symptoms and cognitive coping strategies. Correlations between bully victimization total score and depression/ anxiety were .59 and .57, respectively. Bivariate correlations between cognitive coping strategies and depression ranged from .03 (Positive refocusing) to .61 (Rumination). The same pattern was observed for the correlations between cognitive coping strategies and anxiety symptoms, ranging from .01 (Positive Refocusing) to .55. The highest Pearson correlations (between .45 and .61) were found between cognitive coping strategies rumination, catastrophizing, and self-blame and both types of emotional problems: depression and anxiety. Other significant, but lower correlations with depression and anxiety were found for other-blame and acceptance (between .31 and .35). Although the correlations between planning and depression/ anxiety also were significant (.16 and .14, respectively), they only explained a marginal amount of the variance. Also the Pearson correlations among cognitive coping strategies can be found in Table 3. Correlations ranged from .16 (catastrophizing and positive refocusing) to .68 (self-blame and rumination). Although the cognitive coping strategies were significantly related, there was no evidence of logical or statistical problems due to multicollinearity. In general, correlation values of .90 and higher are considered indicative of multicollinearity of the data (Tabachnick & Fidell, 2001). Subsequently, MRA was performed with depressive symptoms as outcome variable (Table 4). First, the background variables gender and ‘number of life events’ were entered into the regression analysis as the first block (method ¼ enter), in order to control for these variables. Subsequently the variable bully victimization total score was entered. The nine cognitive strategies were entered as the third block (method ¼ enter). Finally, the interaction effects between bully victimization and cognitive strategies were added (method ¼ stepwise). The total model explained 63% of the variance of depressive symptoms (p < .001). After partialling out the effects of gender, life events and cognitive strategies, bully victimization remained a significant direct predictor of depressive symptoms (p < .001). With regard to the cognitive strategies, significant (direct) positive
Table 2 Descriptives and alpha reliabilities of depression, anxiety and cognitive coping scales (N ¼ 582).
Depressive symptoms Anxiety symptoms Cognitive coping Selfblame Otherblame Acceptance Rumination Catastrophizing Planning Putting into Perspective Positive Reappraisal Positive Refocusing
M (Sd)
Range
Cronbach's a
21.12 (9.64) 13.67 (5.61)
14e70 10e50
.93 .90
4e19 4e20 4e20 4e20 4e20 4e20 4e20 4e20 4e20
.74 .72 .72 .82 .74 .82 .77 .72 .81
7.51 6.29 8.73 7.89 6.33 9.68 9.72 9.23 9.83
(3.01) (2.47) (3.39) (3.52) (2.88) (3.78) (3.90) (3.57) (4.10)
1158
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
Table 3 Correlations among study variables (N ¼ 582).
1 Bully victimization 2 Depression 3 Anxiety Cognitive coping 4 Selfblame 5 Otherblame 6 Acceptance 7 Rumination 8 Catastrophizing 9 Planning 10 Putting into Perspective 11 Positive Reappraisal 12 Positive Refocusing
1 r
2 r
3 r
e .59*** .57***
e
.36*** .26*** .24*** .42*** .41*** .18*** .04 .11** .03
..49*** .32*** .35*** .61*** .53*** .16*** .05 .04 .03
.87***
4 r
5 r
6 r
7 r
8 r
9 r
10 r
11 r
12 r
e .36*** .64*** .68*** .53*** .51*** .38*** .41*** .24***
e .38*** .47*** .50*** .46*** .31*** .35*** .26***
e .62*** .49*** .58*** .53*** .56*** .45***
e .63*** .52*** .28*** .36*** .21***
e .35*** .23*** .30*** .16***
e .56*** .70*** .54***
.67*** ..57***
e .60***
e
e .45*** .33*** .31*** .55*** .52*** .14** .04 .04 .01
***: p < .001; **: p < .01; *: p < .05.
‘predictors’ of depressive symptoms were self-blame, rumination and catastrophizing (higher use of these strategies related to more depressive symptoms). Significant (direct) negative ‘predictors’ were planning and positive reappraisal (see Table 4). The first significant interaction effect that was found was: being bullied by rumination. To understand this effect the sample was divided in three equal sized groups on basis of their rumination scores (low, medium, high). In the low rumination group the Pearson correlation between bully victimization and depressive symptoms (corrected for gender and life events) was .20. In the medium rumination group the Pearson correlation was .35 and in the high rumination group the correlation was .55, showing that higher rumination strengthened the association between bully victimization and depressive symptoms. Also a significant (negative) interaction effect was found for being bullied by positive refocusing. Again, the sample was divided in three equal sized groups on basis of their positive refocusing scores. In the low positive refocusing group the correlation (corrected for gender and life events) was .60, in the medium positive refocusing group the correlation was .44 and in the high positive refusing group the correlation was .38. This indicated that higher positive refocusing lowered the association between bully victimization and depressive symptoms. Next, MRA was performed with anxiety symptoms as dependent variable (Table 4). After controlling for gender and life events in the first block, and entering the bully victimization total score in the second block, the nine cognitive strategies were entered. Direct, positive significant effects were found for self-blame, other-blame, rumination, and catastrophizing, whereas negative significant effects were found for planning and positive reappraisal. In the next block, the interaction effects were added (stepwise). The total model explained 58% of the variance of depressive symptoms (p < .001). Significant interaction were found for bully victimization with rumination, catastrophizing and positive reappraisal.
Table 4 MRA on Depression and Anxiety, with interaction terms: method enter. Depression
b Background variables Gender Life events Bully victimization Cognitive coping Selfblame Otherblame Acceptance Rumination Catastrophizing Planning Putting into Perspective Positive Reappraisal Positive Refocusing Significant moderators Bullied by rumination Bullied by Positive Refocusing Bullied by Catastrophizing Bullied by Positive Reappraisal Model Explained variance (R2) ***: p < .001; **: p < .01; *:p < .05.
Anxiety
b
t
t
.09 .21 .22
3.41** 7.24*** 6.19***
.07 .24 .21
2.34* 7.70*** 5.02***
.14 .06 .02 .28 .09 .09 .01 .17 .03
3.52*** 1.87 0.47 6.24*** 2.52* 2.12* 0.34 4.00*** 0.85
.14 .10 .02 .22 .08 .10 .01 .16 .02
3.40** 2.75** 0.41 4.42*** 1.99* 2.27* 0.30 3.40** 0.65
.13 .08
3.99** 2.95**
F(14,526) ¼ 67.94; p ¼ .000 R2 ¼ .63
.09
2.25*
.14 3.34** .10 3.04** F(15,561) ¼ 51.24; p ¼ .000 R2 ¼ .58
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
1159
With regard to the interpretation of the interaction effect of bully victimization and rumination: the sample was divided in three equal sized groups on basis of their rumination scores (low, medium, high). The following correlations were found in the low, medium and high rumination groups (after correction for life events and gender): .24, .37 and .44, showing that higher rumination also strengthened the association between bully victimization and anxiety symptoms. With regard to the interaction effect of bully victimization and catastrophizing, again the sample was divided in three equal sized groups (on basis of catastrophizing scores). The sample correlations were .21, .47, and .46 in the low, medium and high score groups, respectively, showing that both medium and high catastrophizing strengthened the association between bully victimization and anxiety symptoms. With regard to positive reappraisal, three equal sized groups were also created. Correlations between bully victimization and anxiety scores were .57 in the low group, .52 in the medium group, and .42 in the high score group. These results indicated that higher positive reappraisal lowered the association between bully victimization and anxiety. Discussion The present study focused on the role that specific cognitive coping strategies played in the relationship between bully victimization and symptoms of depression and anxiety. First, a strong direct relationship was found between bully victimization and both depression and anxiety symptoms, confirming results of previous studies (Hawker & Boulton, 2000). On top of this direct effect of bullying, specific cognitive coping strategies added a significant amount to the explained variance of depression and anxiety. The strongest direct positive effects were found for self-blame and rumination, whereas the strongest negative effects was found for positive reappraisal. No significant effect was found for Catastrophizing in the regression analysis, however a strong Pearson correlation was found between Catastrophizing and depression/anxiety. These results confirmed the findings of previous studies suggesting that by using cognitive styles such as rumination, catastrophizing and self-blame people may be more vulnerable to emotional problems than others, while by using other styles, such as positive reappraisal people may be more resilient (eg., Garnefski et al., 2001). As far as we know, no studies had been performed before that focused on the moderating role that specific cognitive coping strategies might play in the relationship between bully victimization and symptoms of depression. The present study filled in this gap by identifying two moderators in the relationship between bullying and depression. The first moderator was rumination. Using this strategy appeared to strengthen the effect of bullying on depression. The second moderator was positive refocusing. This strategy appeared to reduce the effect of bullying on depression. Coping strategies that moderated the effect of bullying on anxiety symptoms were rumination, catastrophizing (strengthening the effect on anxiety) and positive reappraisal (reducing the effect on anxiety). Before discussing the possible implications of the results, it is important to address some limitations. One important issue was that the detection of psychological distress as well as the assessment of bully victimization and cognitive strategies was made on basis of self-report. In order to be able to draw firm conclusions about cognitive coping strategies contributing to risk and resilience of emotional problems after bully victimization, further studies are necessary. In such studies more qualitative methods should be applied, for example by using in depth interviews and including narrative measures or daily diaries. In addition, all students of our sample were attending schools for intermediate vocational education. Another issue is that the results of the present study are based on cross-sectional data. Therefore, it is important to acknowledge that no conclusions can be drawn about causality or directions of influence. For purposes of generalizability, the study should be repeated with other samples and with other methods, while prospective elements should be included. If the results of our study can be confirmed in further studies, they might have important practical implications. Firstly, assessment of maladaptive cognitive coping strategies might help to identify adolescents with increased risk of developing symptoms of depression and/or anxiety. Secondly, and even more importantly, the results might provide targets for intervention. One of the important conclusions that was already drawn in the review study of Greenberg, Domitrovich and Bumbarger (2001) concerning the prevention of mental disorders in school-aged children, was that (preventive) interventions should preferably be focused at risk and protective factors rather than to categorical problem behaviors. During the past decades (including the present study) bully victimization has been clearly found to place an adolescent at increased risk for psychopathology (Hawker & Boulton, 2000). The present study could add to this conclusion that on top of that certain maladaptive cognitive strategies might increase the risk for psychopathology in adolescents, whereas other cognitive strategies might be able to reduce the risk. An important target for interventions might therefore be to prevent maladaptive cognitive coping strategies from turning into long-established and difficult-to-change styles by reducing non-adaptive strategies and acquiring more adaptive strategies. The results of the present study might give specific suggestions for the content of such interventions. Firstly, when helping an adolescent to cope with bully victimization, it could be important for health care professionals to assess to what extent the adolescent uses self-blaming, rumination, or catastrophizing in response to bully victimization. Secondly, the adolescent could be assisted to become more aware of and to challenge these maladaptive cognitive coping patterns and to acquire new, more adaptive strategies such as positive refocusing or positive reappraisal. Although specific interventions on changing cognitive coping strategies are not available yet, these could easily be integrated in existing Cognitive Behavioral Therapies, which are widely used and evidence based (Hofmann, Asnaani, Vonk, Sawyer, & Fang, 2012). However, as mentioned before, before drawing firm conclusions, further studies in other samples and with other research methods that focus on the same research questions, are necessary. If the results are confirmed, they carry important
1160
N. Garnefski, V. Kraaij / Journal of Adolescence 37 (2014) 1153e1160
implications for the focus and content of intervention and prevention of mental health problems after the experience of bullying by adolescents. Acknowledgments The authors would like to thank the students who participated in this research project: David Kooij, Rob Krens, and Birgit Olsen. References Arrindell, W. A., & Ettema, J. H. M. (1986). SCL-90, Handleiding bij een multidimensionele psychopathologie-indicator [SCL-90, Manual for a multidimensional psychopathology indicator]. Lisse: Swets & Zeitlinger B.V. Bitsch Hansen, T., Steenberg, L. M., Palic, S., & Elklit, A. (2012). A review of psychological factors related to bullying victimization in schools. Aggression and Violent Behavior, 17, 383e387. Copeland, W. E., Wolke, D., Angold, A., & Costello, J. (2013). Adult Psychiatric outcomes of bullying and being bullied by peers in Childhood and Adolescence. JAMA Psychiatry, 70, 419e426. Craig, W., Harel-Fisch, Y., Fogel-Grinvald, H., Dostaler, S., Hetland, J., Simons-Morton, B., &, HBSC Bullying writing group. (2009). A cross-national profile of bullying and victimization among adolescents in 40 countries. International Journal of Public Health, S216eS224. Derogatis, L. R. (1977). SCL-90: Administration, scoring and procedures manual-I for the r(evised) version. Baltimore: John Hopkins University School of Medicine, Clinical Psychometrics Research Unit. Garnefski, N., & Kraaij, V. (2006). Relationships between cognitive emotion regulation strategies and depressive symptoms: a comparative study of five specific samples. Personality and Individual Differences, 40, 1659e1669. Garnefski, N., Kraaij, V., & Spinhoven, P. (2001). Negative life events, cognitive emotion regulation and emotional problems. Personality and Individual Differences, 30, 1311e1327. Garnefski, N., Kraaij, V., & Spinhoven, P. (2002). CERQ: Manual for the use of the cognitive Emotion Regulation Questionnaire: A questionnaire measuring cognitive coping strategies. Leiderdorp, The Netherlands: DATEC. Garnefski, N., Kraaij, V., & Van Etten, M. (2005). Specificity of relations between adolescents’ cognitive emotion regulation strategies and internalizing and externalizing psychopathology. Journal of Adolescence, 28, 619e631. Greenberg, M. T., Domitrovic, C., & Bumbarger, B. (2001). The prevention of mental disorders in school-aged children: current state of the field. Prevention & Treatment, 4. article 1. Hampel, P., Manhal, S., & Hayer, T. (2009). Direct and relational bullying among children and adolescents: coping and psychological adjustment. School Psychology International, 30, 474e490. Hawker, D. S. J., & Boulton, M. (2000). Twenty years' research on peer victimization and psychosocial maladjustment: a meta-analytic review of crosssectional studies. Journal of Child Psychology and Psychiatry and Allied Disciplines, 41, 441e455. Hofmann, S. G., Asnaani, A., Vonk, I. J. J., Sawyer, A. T., & Fang, A. (2012). The efficacy of cognitive behavioral therapy: a review of meta-analyses. Cognitive Therapy Research, 36, 427e440. Kochenfelder-Ladd, B., & Skinner, K. (2002). Children's coping strategies: moderators of the effects of peer victimization? Developmental Psychology, 38, 267e278. Kyriakides, L., Kaloyirou, C., & Lindsay, G. (2006). An analysis of the Revised Olweus Bully/Victim Questionnaire using the Rasch measurement model. British Journal of Educational Psychology, 76, 781e801. Nakamoto, J., & Schwartz, D. (2010). Is peer victimization associated with academic achievement? A meta-analytic review. Social Development, 19, 221e242. Olweus, D., & Limber, S. P. (2010). Bullying in school: evaluation and dissemination of the Olweus Bullying Prevention Program. American Journal of Orthopsychiatry, 80, 124e134. Silberschmidt Viala, E. (2014). The fighter, the punk, and the clown: how to overcome the position of victim of bullying. Advance online publication Childhood. http://dx.doi.org/10.1177/0907568214521845. Solberg, M. E., & Olweus, D. (2003). Prevalence estimation of school bullying with the Olweus Bully/Victim Questionnaire. Aggressive Behavior, 29, 239e268. Tabachnick, B. G., & Fidell, L. S. (2001). Using multivariate statistics (4th ed.). Needham Heights, USA: Allyn & Bacon. Tenenbaum, L. S., Varjas, K., Meyers, J., & Parris, L. (2011). Coping strategies and perceived effectiveness in fourth through eighth grade victims of bullying. School Psychology International, 32, 263e286. Thompson, R. A. (1991). Emotional regulation and emotional development. Educational Psychology Review, 3, 269e307. Ttofi, M. M., & Farrington, D. P. (2011). Effectiveness of school-based programs to reduce bullying: a systematic and meta-analytic review. Journal of Experimental Criminology, 7, 27e56.
