Novin, F.S.
Citation
Novin, F. S. (2011, June 16). Moroccan-Dutch adolescents' emotional functioning : in between cultures?. Retrieved from https://hdl.handle.net/1887/17713
Version: Not Applicable (or Unknown)
License: Licence agreement concerning inclusion of doctoral thesis in the Institutional Repository of the University of Leiden Downloaded from: https://hdl.handle.net/1887/17713
Note: To cite this publication please use the final published version (if applicable).
Emotion Awareness and Social Functioning Related to Internalizing Symptoms in Bicultural
and Monocultural Groups
Evidence explaining adolescents’ internalizing symptoms suggests influences of social as well as emotional competence. However, these competencies are interrelated. In the present study we examined the unique contribution of emotion awareness to the prediction of emotional difficulties and social anxiety, over and above indices of social functioning (trust, pro-social behavior, and peer problems). Various questionnaires were administered among 612 Dutch, Moroccan, and Moroccan-Dutch 12 to 16-year-old adolescents. Using hierarchical regression analyses, our findings show that social functioning, particularly having peer problems, as well as understanding of own and others’ emotions (emotion awareness) uniquely contribute to the prediction of internalizing symptoms. We argue that these two constructs are relatively independent, and taken together they have a cumulative contribution to the prediction of internalizing symptoms. In addition to these results applicable to all cultural groups, culture- specific outcomes are discussed in light of the role of emotions in diverse cultures.
Novin, S. & Rieffe, C. (under review). Emotion awareness and social functioning related to internalizing symptoms in bicultural and monocultural groups.
Introduction
Developmental psychologists have been intrigued by the interplay between youngsters’ socio-emotional functioning and psychopathology for many years in order to predict, prevent, and treat mental health problems in youngsters. Although more attention has been given to the more noticeable, social part of children’s and adolescents’ functioning, there is a growing recognition of the importance of the less visible emotional part of functioning. Across various cultures, studies show that emotion awareness - the ability to identify and analyze one’s own and others’ emotions and to appreciate their value- contributes to fewer internalizing problems, such as depression, worry, somatic complaints, and social anxiety (e.g., Rieffe, Oosterveld, & Meerum Terwogt, 2006; Rieffe et al., 2010). However, the contributions of social and emotional functioning to youngsters’ internalizing symptoms are usually examined separately and studies have typically focused on one cultural group. The aim of this study was to examine the unique contribution of emotion awareness on internalizing symptoms over and above social functioning in monocultural and bicultural youth.
Internalizing symptoms and social competence
Children’s and adolescents’ internalizing symptoms are strongly related to impairments in social functioning. Children who have peer difficulties are less friendly and more socially inhibited, and they experience higher levels of emotional difficulties and social anxiety, whereas interacting sociable with peers is related to fewer internalizing problems (e.g., Chen et al, 2002; Cole, Martin, Powers, & Truglio, 1996; Kiesner, 2002; Nangle, Erdley, Newman, Mason, & Carpenter, 2003). Rotenberg and colleagues’ study (2005) focusing on a specific aspect of social functioning, trustworthiness in others, demonstrated that children who held lower levels of trust in peers reported substantially more depressive symptoms. Additionally, a large body of research shows that higher numbers of and quality of friendships are also associated with healthy psychological wellbeing (Ladd, 1999;
Parker, Rubin, Erath, Wojslawowicz, & Buskirk, 2006). In fact, several longitudinal studies indicate the significance of children’s social functioning for internalizing problems over time. For example, Burt, Obradovi!, Long, and Masten (2008) demonstrated that social functioning of middle school-aged children predicts anxious-depressive symptoms and somatization in adolescence and adulthood. Clearly, these findings indicate that children’s functioning in relation to others, contributes to their intrapsychological wellbeing.
There are several reasons why children’s social functioning can contribute to internalizing problems. On the positive side,
acting pro-socially by being kind and helping others offers opportunities to receive positive attention that can foster increases in self-esteem and positive mood (Laible, Carlo, &
Roesch, 2004; Martin & Huebner, 2007). The ability to trust others, for example believing that others will provide you help and protection when needed, is also likely to contribute positively to a person’s wellbeing, since it reflects a positive world-view where a child feels secure enough to explore his or her environment without fear of making mistakes that will have severe negative consequences. On the negative side, peer problems, such as being lonely, rejected or bullied, undermine feelings of self-worth and lead to fearing social interactions (Hawker & Boulton, 2000;
Storch & Masia-Warner, 2004).
Internalizing symptoms and emotion awareness
Another important component of healthy psychological functioning is emotion awareness. Emotion awareness can be described as the attentional process aimed at monitoring and differentiating one’s own emotions, identifying their antecedents, and communicating them with others, whereas little attention is paid to the physical arousal that is part of the emotion experience (Rieffe, Oosterveld, Miers, Meerum Terwogt, & Ly, 2008).
Additionally, emotion awareness also comprises attitudinal constructs, such as how emotion experiences are valued and should be expressed.
In particular, the ability to differentiate discrete emotions is known to be strongly related to better emotion regulation (Barrett, Gross, Christensen, & Benvenuto, 2001) and fewer symptoms of mental health problems (Taylor & Bagby, 2000).
Several studies confirm that also in elementary school age children, a lower level of emotion awareness is related to more internalizing problems (Rieffe, et al. 2006, 2007b, 2008, 2009, 2010). Studies have consistently shown that there are three major aspects of emotion awareness (i.e. the inability to differentiate between emotions and talk about them, and a stronger awareness of bodily symptoms during an emotion experience) that are unique contributors to the prediction of various internalizing symptoms. However, a stronger tendency to appreciate one’s own and/or others’ emotions contributed to more social anxiety, but to fewer symptoms of depression (Rieffe et al, 2007a; 2008; submitted), showing differential associations for social anxiety and depression with some of the attitudinal aspects of emotion awareness.
Social functioning and emotion awareness
These studies clearly demonstrate contributions of both social functioning and awareness of one’s own and others’ emotions to internalizing symptoms in childhood. However, social functioning
and emotion awareness do not seem to be entirely independent constructs (Rose-Krasnor, 1997; Denham, 1998). For example, developmental scholars acknowledge that understanding emotions is an important predictor of children’s social behavior (e.g., Eisenberg et al., 1996; Denham et al., 2003; Saarni, 1999). That is, recognizing emotions from facial expressions, behavioral cues, and social contexts allows people to react responsively and thus enhances social relationships. Earlier studies have shown that children who show a better understanding of others’ emotions behave more pro-socially to peers and are rated as more likable by peers and more socially competent by teachers (Denham, 1986; Denham, McKinley, Couchoud, & Holt, 1990). Furthermore, children’s emotion knowledge predicts social competence longitudinally (Denham et al., 2003). Given the relatedness of social functioning and emotion awareness, the question of their unique contributions to youngsters’ internalizing problems arises.
Furthermore, with the exception of emotion knowledge, other aspects of emotion awareness, such as analyzing one’s own emotions, verbally sharing emotions, and attending to others’
emotions, and their relation to youngsters’ social functioning are under-researched. In order to fully understand children’s psychological wellbeing, it is important to consider how the diverse components of emotion awareness impact youngsters’
social functioning and, in turn, their experience of internalizing problems.
The present study
The aim of the present study was to examine the relationships between emotion awareness, social functioning, and internalizing symptoms in a population of adolescents by means of self-report questionnaires. The age range of 12 – 16 years old was chosen because at this age teenagers are increasingly able to reflect upon their internal states and emotions (Harris, 1989), allowing the use of self-reports regarding own emotional and social functioning.
Moreover, we included two monocultural groups with distinct cultural backgrounds (Moroccan and Dutch) and one bicultural group that is influenced by both these cultures (Moroccan-Dutch) in order to increase the external validity of this study.
We first examined associations between the key aspects of emotion awareness that were identified by Rieffe and colleagues (2007a; 2008) (differentiating emotions, verbal sharing of emotions, not hiding emotions, bodily awareness, attending to others’ emotions, and analyses of emotions) and three indices of social functioning (trust in others, pro-social behavior, and peer problems). It was expected that components of emotion awareness concerning the display of one’s emotion (verbal sharing and not hiding emotions) would place oneself in a vulnerable position and therefore would be related to more trust
in others. Understanding one’s own emotions and interest in others’ emotions (analyses of and attending to emotions) were expected to be associated with more pro-social behavior and fewer peer problems.
Second, the extent to which aspects of social functioning are associated with and contribute to the prediction of two indices for internalizing symptoms (emotional difficulties and social anxiety) was examined. We predicted that peer problems would be positively associated with and contribute to the prediction of internalizing symptoms, whereas this relationship was expected to be reversed for trust and pro-social behavior.
Next, we assessed which aspects of emotion awareness contribute to internalizing symptoms over and above social functioning using hierarchical regression. Therefore, we first added the three social indices to the regression analysis, and, in the second step, added the six key aspects of emotion awareness.
Based on previous findings, we expected that three aspects of emotion awareness, an inability to differentiate between emotions, difficulty talking about emotions, and a stronger tendency to attend to bodily symptoms that are part of the emotion experience, would significantly contribute to the prediction of self-reported internalizing symptoms.
Additionally, we predicted that various aspects of emotion awareness could have differential relations with depression/worry and social anxiety symptoms. Although both are facets of internalizing problems, they each have unique characteristics that may be associated differently with emotional components (e.g., Wright, Banerjee, Hoek, Rieffe & Novin, 2010). Depressive symptoms and worry are characterized by feeling hopelessness in a general sense, resulting in lack of interest about others and feeling unmotivated to engage in interpersonal activities. Social anxiety however revolves around the specific fear of doing and saying something that will result in embarrassment or humiliation in presence of others. Whereas persons with depression have a lack of interest in others, people with social anxiety are very much focused on others. With respect to the relation between these symptoms and emotion awareness, we therefore predicted that a stronger willingness to attend to and understand one’s own and/or other people’s emotions would contribute to fewer emotional difficulties in terms of depression and worry, but to more social anxiety.
Although the concept of emotion awareness appears to be universal, key aspects influencing adolescents’ psychological health may differ cross-culturally due to cultural differences in beliefs and values concerning emotions. In line with the well- known individualistic-collectivistic dimension (Triandis, 1995), attention to internal processes, such as emotion experiences, might be less common within collectivistic-oriented (e.g.,
Moroccan) than in individualistic-oriented cultures (e.g., Dutch), because it emphasizes personal instead of social processes.
Indeed, cross-cultural studies show that parents’ emotion socialization patterns in collectivistic-oriented cultures are more characterized by focusing on ‘what is appropriate’ and on the social situation, whereas parents in individualistic-oriented cultures are more likely to focus on internal processes related to emotion, for example the causes of emotion (e.g., Eisenberg et al., 1996; Wang & Fivush, 2005). Consequently, aspects of emotion awareness that refer to awareness of own emotions are more likely to play significant role in adolescents’ social functioning and psychological wellbeing in individualistic-oriented rather than in collectivistic-oriented cultures. Conversely, aspects of emotion awareness that reflect awareness of other’s emotions may be more important in collectivistic-oriented cultural groups.
However, given the exploratory nature of this cross-cultural study, no specific hypotheses could be formulated based on existing literature.
Method Participants
As illustrated in Table 1, the participants were 612 adolescents from three cultural groups (Dutch, Moroccan, and Moroccan- Dutch). Overall, 49% of the participants were male. The three groups did not differ in age.
The Dutch and Moroccan-Dutch adolescents were recruited through their schools in the Randstad, the highly urbanized and industrialized part of the Netherlands with the highest concentration of people of Moroccan origin. These adolescents received schooling at preparatory secondary vocational or intermediate level. The adolescents and both parents in the Dutch group were all born in the Netherlands. The majority reported not to be religious (62%) and to speak only Dutch at home (96%).
The Moroccan-Dutch adolescents had at least one parent who was born in Morocco. Eight percent of the Moroccan-Dutch adolescents had one parent who was born in the Netherlands (8%). These parents did have a Moroccan background. The Moroccan-Dutch adolescents themselves were so-called second generation: they were all born in the Netherlands or had immigrated to the Netherlands before the age of 6. All Moroccan- Dutch adolescents reported being Muslim. Although in more than half of the families (54%) two languages were spoken (Dutch and Arabic or Berber) and 22% reported speaking only Arabic or Berber at home, all Moroccan-Dutch adolescents were fluent in Dutch.
All Moroccan adolescents lived in and were born in Morocco, as well as their parents. They all reported being Muslim and
speaking Arabic or Berber at home. Compared to the Dutch adolescents, the Moroccan-Dutch and Moroccan participants came from larger families (average number of siblings 1.7 versus 3.3 and 3.5).
Table 1
Sample Demographics
Instruments
Internalizing Symptoms. The Strengths & Difficulties Questionnaire (SDQ) is a brief behavioral screening questionnaire (Goodman, 1997; Dutch translation by Van Widenfelt, Goedhart, Treffers, & Goodman, 2003; Arabic translation retrieved from sdqinfo.org), measuring five aspects of adolescents’ psychological functioning. One of these aspects, which is reflected in the scale Emotionality (5 items), assesses participants’ emotional difficulties, including depressive-anxious symptoms and worry.
Participants can rate each item on a three-point scale (0 = not true, 1 = somewhat true, 2 = certainly true). Item examples are:
‘I worry a lot’ and ‘I am often unhappy, down-hearted or tearful’.
The coherence between these symptoms is reflected in the good internal consistency (Table 1).
The Social Anxiety Scale for Adolescents (SAS-A) (La Greca &
Lopez, 1998) is a modified version of the Social Anxiety Scale for Children—Revised (SASC-R). A Dutch version of the questionnaire was used for this study (Dekking, 1983). The SAS-A reflects three subscales: Fear of Negative Evaluation, Social Avoidance and Distress in General, and Social Avoidance Specific to New Situations or Unfamiliar Peers, and consists of 18 items which are scored on a five-point scale according to how true the item is (1 = not at all, 5 = all the time). Examples of items are: ‘I only talk to people I known really well’ and ‘I worry about what others think of me’. Items that are formulated positively are reversed scored.
The questionnaire showed good internal consistency in previous studies (Meerum Terwogt, Rieffe, Miers, Jellesma, & Tolland, 2006), which was confirmed in this study (Table 2).
N Age
Cultural group
Total Male Female M SD Range
Dutch 282 149 133 14 years,
7 months 1 year,
4 months 12-16 years Moroccan
-Dutch
206 92 114 14 years, 10 months
1 year, 3 months
12-16 years Moroccan 124 58 66 14 years,
7 months 1 year,
1 month 12-16 years
Table 2
Psychometric Properties of the Emotionality, Social Anxiety, Social Functioning, and EAQ
Dutch Moroccan -Dutch
Moroccan No.
of items
!
M (SD) M (SD) M (SD)Emotionality 5 .71 0.512
(0.46) 0.323
(0.35) 0.701 (0.47) Social Anxiety 18 .92 2.042
(0.62) 1.843
(0.70) 2.661 (0.58) Social Functioning
Trust 12 .77 2.531
(0.58) 2.521
(0.70) 2.332 (0.65) Pro-social Behavior 5 .65 1.471
(0.38) 1.521
(.41) 1.521 (0.36) Peer problems 5 .29 0.363
(0.31)
0.442 (0.31)
0.691 (0.34) EAQ Scales
Differentiating 7 .78 2.452 (0.41)
2.561 (0.38)
2.033 (0.39) Verbal Sharing 3 .64 2.201
(0.55) 2.321
(0.58) 1.952 (0.52) Not Hiding Emotions 5 .70 2.031
(0.50)
2.061 (0.50)
1.802 (0.43) Bodily Awareness 5 .66 1.9712
(0.53) 2.051
(0.45) 1.892 (0.48) Others’ Emotions 5 .71 2.501
(0.45)
2.382 (0.48)
2.382 (0.43) Analyses of Emotions 5 .74 2.042
(0.49)
2.052 (0.52)
2.541 (0.38) Note. Number subscripts indicate cultural group differences at p< .01 Social Functioning. Two scales of the above-described SDQ assessing adolescents’ social functioning were used in this study:
Peer Problems (5 items) and Pro-Social Behavior (5 items). The scoring is reversed for items which were formulated positively, for example “I am generally liked by other children”. The internal consistency of the scale Pro-Social Behavior is usually good, but the internal consistency of the scale Peer Problems is known to be low (Van Widenfelt, et al., 2003). This was confirmed in this study (Table 2).
The Trust Questionnaire is an adaptation of Rotter’s Interpersonal Trust Scale (1967) to assess the degree to which adolescents have a general trust in other people. The adapted
scale consists of 12 items in which participants indicate on a five- point scale how true the statements are (0 = not true at all, 4 = totally true). Examples of items are ‘most people tell the truth’
and ‘most people provide you help when you need it’. Half of the items are negatively formulated and are thus reversed-scored.
Internal consistency was satisfactory (Table 2).
Emotion Awareness. The Emotion Awareness Questionnaire (EAQ) (Rieffe et al, 2007a; 2008) consisting of 30 items, was designed for children over 8 years old with a 6-factor structure describing 6 key aspects of emotion awareness: (1) Differentiating Emotions, (2) Verbal Sharing of Emotions, (3) Not Hiding Emotions, (4) Bodily Awareness of Emotions, (5) Attending to Others’ Emotions, and (6) Analyses of (Own) Emotions. Twenty items are negatively formulated and thus reversed-scored.
Respondents are asked to rate the degree to which each item is true about them on a three-point scale (1= not true, 2 = sometimes true, 3 = often true). In all the scales, a higher score represented a higher presence of emotion awareness, with the exception of Bodily Awareness, in which a higher score indicates lower attention to bodily symptoms. All scales showed good internal consistencies in other studies (Cronbach’s Alpha’s ranging between .74 and .77; Rieffe et al, 2008; 2009), which was confirmed in this study (Table 2).
Translation of the questionnaires
The SDQ had already been translated and validated for previous studies among Arabic populations (e.g., Thabet, Stretch, &
Vostanis, 2000; Alyahri & Goodman, 2006). We translated the Trust Questionnaire, the SAS-A, and the EAQ for the Moroccan adolescents in Morocco by back-translation, as suggested by Brislin, Lonner, and Thorndike (1973). For the translation, two translators translated the questionnaires from Dutch into Arabic.
Two other bilingual translations translated the Arabic questionnaires back into Dutch. One of the translators has a Ph.D.
in Arabic languages. Differences in the original and back- translated versions were discussed by the translation team and resolved through joint agreement.
Procedure
The procedure was similar in the Netherlands and Morocco. The questionnaires were handed out to adolescents in classrooms during school hours along with other questionnaires that are excluded in this study. Prior to the assessment, the aims of the study were explained, and adolescents were assured that participation was voluntary and anonymous. The administrator gave instructions for each questionnaire separately and stayed in the class in order to answer questions. It took approximately 30
minutes to complete all questionnaires. In the Netherlands, parental permission for adolescents younger than 16 years was obtained. In Morocco, official governmental permission was obtained to conduct the study at the selected schools.
Results Correlations
Pearson correlations between the scales were calculated for each cultural group separately (Table 3). The results show that the internalizing symptoms, Emotionality and Social Anxiety, were positively related for each cultural group. The indices of social functioning also were associated with each other. Trust was positively associated with Pro-Social Behavior. Trust was also negatively related to Peer Problems, except for the Moroccan group. More Pro-Social Behavior was correlated with fewer Peer Problems, except for the Moroccan-Dutch group.
The correlations between the six EAQ scales are in line with previous studies (e.g., Rieffe et al., 2008). For all cultural groups, Not Hiding Emotions was positively related to Verbal Sharing of Emotions and Attending to Others’ Emotions. Not Hiding Emotions was also positively associated with Differentiating Emotions, except for the Moroccan group. Similarly, Attending to Others’
Emotion was positively related to Analyses of Emotions, and Bodily Awareness of Emotions was positively associated with Differentiating Emotions, Attending to Others’ Emotions, Analyses of Emotions, and Verbal Sharing of Emotions, again except for the Moroccan group. As noted in the method section, higher scores on the Bodily Awareness scale reflects lower attention to bodily symptoms. Furthermore, Verbal Sharing of Emotions was positively related to Attending to Others’ emotions, except for the Dutch group.
With respect to the associations between aspects of emotion awareness and social functioning, it was found that at least half of the relations were significant for the Moroccan-Dutch and the Dutch group, in contrast to the Moroccan group where two out of eighteen associations were significant. For all cultural groups, Pro- Social Behavior was negatively related to Bodily Awareness of Emotions and positively to Attending to Others’ Emotions. As expected, both Verbal Sharing of Emotions and Not Hiding Emotions were related to more Trust, except for the Moroccan group. Furthermore, Analyses of Emotions was associated with more Pro-Social Behavior in most groups. Analyses of Emotions and Attending to Others’ Emotions were, however, only related to fewer Peer Problems for the Dutch group. In contrast, Differentiating Emotions was related to fewer Peer Problems and to higher Trust levels for the Dutch and Moroccan-Dutch group.
Next, correlations between the indices of social functioning and internalizing problems were conducted. As expected, Trust levels were negatively associated with Emotionality and Social Anxiety symptoms, except for the Moroccan group. Moreover, Peer Problems were positively related to Emotionality and Social Anxiety symptoms. Contrary to our expectations, Pro-Social Behavior was not related to internalizing symptoms.
Consistent with previous research, various aspects of emotion awareness were negatively related to both Emotionality and Social Anxiety, namely Differentiating Emotions and Bodily Awareness of Emotions for all cultural groups and Verbal Sharing of Emotions, Not Hiding Emotions, and Bodily Awareness of Emotions for the Dutch and Moroccan-Dutch group. Also consistent with previous work, we found that Attending to Others’ Emotions was positively related to Emotionality for the Dutch group.
Table 3
Correlations between Emotionality, Social Anxiety, Social Functioning and the Scales of the EAQ
* p < .05; ** p < .01; *** p <.001 Regression
Next, we examined which aspects of emotion awareness independently influenced the level of adolescents’ self-reported Emotionality over and above that of the three indices for social functioning. A hierarchical regression analysis was conducted with Emotionality as the criterion variable, in which Gender, Trust, Pro- Social Behavior, and Peer Problems were entered in the first step and the six EAQ scales in the second. The standardized beta
coefficients are displayed in Table 4 for each cultural group separately.
Table 4
Summary of Hierarchical Regression Analysis for Social Functioning and Emotion Awareness on Emotionality for Dutch, Moroccan-Dutch and Moroccan Adolescents Separately
Dutch Moroccan-Dutch Moroccan Variable R2adj ! p R2adj ! p R2adj ! p Step 1:
.29 .12 .10
Gender .29 .000 .05 .072 .25 .004
Social Functioning
Trust -.24 .000 -.22 .001 .03 .756
Pro-social behavior .14 .012 .14 .059 .06 .545 Peer problems .35 .000 .23 .001 .27 .003 Step 2:
.52 .26 .21
Gender .14 .004 .07 .303 .25 .004
Social Functioning
Trust -.15 .001 -.14 .039 .09 .307
Pro-social behavior .04 .261 .05 .532 -.08 .408 Peer problems .22 .000 .17 .010 .18 .035
EAQ scales
Differentiating -.31 .000 -.27 .000 -.29 .001 Verbal Sharing -.18 .000 -.01 .844 -.11 .182 Not Hiding .04 .461 -.07 .365 -.04 .663 Bodily Awareness -.25 .000 -.20 .010 -.21 .014 Others’ Emotions .07 .224 .05 .533 .12 .206 Analyses of Emotions -.10 .048 -.08 .279 -.14 .085
An independent positive effect was found for Gender for the Dutch and Moroccan group, indicating that being a Dutch or Moroccan girl is related to higher Emotionality scores, thus to more emotional difficulties. In addition, Peer Problems contributed positively to the prediction of Emotionality in all three cultural groups, whereas Pro-Social Behavior also contributed positively in this prediction, but only for the Dutch group. Trust was related to
lower Emotionality scores for the Dutch and the Moroccan-Dutch groups.
In all cultural groups, the explained variance significantly increased with the addition of the EAQ factors. All indices of social functioning maintained their unique contribution to the prediction of Social Anxiety, except for Pro-Social behavior. As predicted, Differentiating and Bodily Awareness contributed negatively to the prediction of Emotionality. However, the ability to talk about emotions (Verbal Sharing) and the willingness to address them (Analyses of one’s Own Emotions) contributed negatively for the Dutch group only. The total model explained 43% of the variance of adolescents’ self-reported emotional difficulties.
The hierarchical regression with a more specific kind of psychological impairment, Social Anxiety, as a criterion variable showed that Peer Problems significantly contributed positively to the prediction of Social Anxiety in step 1, whereas this was reversed for Trust (for the Dutch and the Moroccan-Dutch group) (Table 5). Moreover, higher levels of Pro-Social Behavior were also associated with more Social Anxiety symptoms for the Dutch group.
Again, a significant increase in the amount of explained variance occurred with the addition of the EAQ scales in step 2 for all three cultural groups. Peer Problems and Trust maintained their unique contribution to the prediction of Social Anxiety, except for the Moroccan-Dutch group. For the Dutch group, both Verbal Sharing of Emotions and Attending to Others’ Emotions predicted a lower level of Social Anxiety. Conversely, for the Moroccan sample Attending to Others’ Emotions contributed positively to the prediction of Social Anxiety. Additionally, Differentiating and Bodily Awareness contributed again to the prediction of Social Anxiety, but now only for the Moroccan and Moroccan-Dutch group. Not Hiding and Analyses of One’s Own Emotions did not contribute for any group. The total model explained 34% of the variance of the self-reported social anxiety in adolescents.
Table 5
Summary of Hierarchical Regression Analysis for Social Functioning and Emotion Awareness on Social Anxiety for Dutch, Moroccan-Dutch and Moroccan Adolescents Separately
Dutch Moroccan-Dutch Moroccan Variable R2adj ! p R2adj ! p R2adj ! p Step 1:
.26 .07 .04
Gender .23 .000 -.02 .800 .04 .658
Social Functioning
Trust -.25 .000 -.22 .002 -.08 .402
Pro-social behavior .12 .042 .02 .789 -.02 .801 Peer problems .35 .000 .16 .026 .24 .010 Step 2:
.34 .21 .12
Gender .22 .000 -.07 .375 -.00 .25 .977
Social Functioning
Trust .20 .000 -.12 .083 -.03 .773
Pro-social behavior .12 .051 -.08 .310 -.15 .124 Peer problems .27 .000 .10 .123 .19 .034
EAQ scales
Differentiating -.10 .107 -.21 .007 -.19 .043 Verbal Sharing -.24 .000 -.13 .096 -.09 .301 Not Hiding .06 .356 -.08 .264 -.07 .434 Bodily Awareness -.12 .056 -.19 .017 -.20 .031 Others’ Emotions .13 .052 .07 .405 .25 .013 Analyses of Emotions -.01 .892 -.02 .779 -.06 .478
Discussion
The aim of the present study was to examine the unique contribution of adolescents’ emotion awareness to the prediction of two internalizing symptoms (emotional difficulties and social anxiety), over and above three indices of social functioning: trust in others, pro-social behavior, and peer problems. The external validity of the results was tested by comparing two distinct monocultural groups (e.g. Dutch and Moroccan) and a bicultural group (e.g. Moroccan-Dutch).
We first examined associations between our three indices of social functioning with different key aspects of emotion awareness. In line with previous studies among Western populations (Denham et al., 1990, 2003; Eisenberg et al., 1996), our findings show that social functioning and emotion awareness are interrelated in expected directions, especially for the Dutch group, but also to a lesser degree for the Moroccan-Dutch group.
As expected, displaying one’s emotions, such as verbally discussing emotion experiences, is related to higher levels of trust for the Dutch and Moroccan-Dutch adolescents. It is likely that people only place themselves in a vulnerable position by expressing internal, private emotions to others when they trust others to react sensitively. This relation might even be stronger in adolescents than in adults and younger children, because feelings of (social) uncertainty are characteristic in this age period (La Greca & Lopez, 1998). In addition, in the Dutch and Moroccan- Dutch samples, adolescents’ ability to identify and differentiate emotions is related to fewer peer problems, whereas appreciating and being willing to address one’s own emotions is associated with more pro-social behavior.
Fewer significant associations between the indices of social functioning and emotion awareness are found for the Moroccan group. Similar to the Dutch and Moroccan-Dutch group, pro-social behavior is related to more bodily awareness and to more interest in others’ emotions. The lack of other associations for the Moroccan group may be explained by the greater influence of behavioral norms in Morocco than in the Netherlands (Gelfand, Bhawuk, Nishii, & Bechtold, 2004). Narrowly defined behavioral norms may leave little room for trust issues and attention to one’s own emotions. Moreover, in the collectivistic-oriented Moroccan culture, attention to one’s own emotions might be less relevant than in individualistic-oriented cultures like the Netherlands, because one’s focus needs to be on social processes rather than internal processes. Our findings indeed show that Moroccan adolescents are more likely to hide their emotions and less likely to talk about them. Consequently, the ability to understand one’s emotions might be less important for social functioning among Moroccan adolescents.
Second, we examined the contribution of three indices of social functioning to the prediction of adolescents’ self-reported internalizing problems: emotional difficulties and social anxiety. In all three groups, high levels of peer problems predict higher emotionality and social anxiety scores. This robust contribution of peer problems across groups, despite its low internal consistency, indicates that the scale measures a broad dimension of peer problems, reflected by items that do not necessarily strongly correspond with one another, yet represent an underlying construct with strong predictive power. Lower levels of
interpersonal trust also predict higher levels of internalizing problems, but only for the Dutch and Moroccan-Dutch group.
Together with the previous results regarding the lack of association between emotion awareness and trust for the Moroccan group, we may conclude that interpersonal trust plays a less important role in Moroccan adolescents’ emotional and psychological functioning than for adolescents in the Netherlands.
It might be worthwhile for future research to explore the underlying reasons for this outcome in adolescents’ attributions and perceptions of social relations.
Unexpectedly, pro-social behavior contributes positively, and not negatively, to both internalizing symptoms for the Dutch group, even though there is a lack of association in the correlation matrix. Post-hoc partial correlations show that the associations between pro-social behavior and internalizing symptoms are mediated by peer problems (p<.05) but not by trust for this group. This suggests that pro-social behavior could backfire, possibly when children are trying too hard to be accepted. Future studies should examine this more closely.
Third, the results show that the addition of our emotion awareness scales into the regression analyses increases the explained variance for all groups substantially. Moreover, the unique contributions of the indices of social functioning to the prediction of internalizing problems remains intact for all groups, especially with respect to peer problems. For the Dutch group, only the contribution of pro-social behavior decreases. For the Moroccan-Dutch group, the contributions of trust and peer problems decrease in the prediction of social anxiety symptoms.
Consistent with the literature (Rieffe et al., 2007, 2008), the ability to differentiate between emotions and higher awareness of bodily symptoms are the strongest predictors for adolescents’
internalizing problems, even over and above indices of social functioning. Additionally, some aspects of emotion awareness show differential links with emotionality (depression/worry) and social anxiety, which could provide us with more insight into the nature of these problems. Although dependent on cultural group as discussed below, trying to understand one’s own emotions contributed to the prediction of lower levels of emotionality, whereas attending and understanding others’ emotions contributed to the prediction of social anxiety. These specific types of emotion awareness may play a mediating factor in the relationship between difficulties in emotion regulation and the internalizing problems at interest (for review see Aldao, Nolen- Hoeksema, & Schweizer, 2010).For example, understanding the external causes of one’s emotions may make it easier to regulate negative emotions adequately, which in turn may decrease depression/worry symptoms. However, given that social anxiety is characterized by high self-awareness, high levels of
understanding one’s emotions may similar effects on social anxiety symptoms.
As noted, unique contributions of aspects of emotion awareness were dependent on cultural group. For the Dutch sample, the ability to verbally share emotions contributes to the prediction of lower levels of social anxiety and emotionality. The importance of verbally sharing emotions seems to be congruent with characteristics of the Dutch culture, where expressing personal needs, such as through emotions, is highly valued and encouraged (Rispens, Hermans, & Meeuws, 1996). Moreover, in Western cultures it may be considered healthier to openly express one’s emotions. Furthermore, in line with previous research, understanding and analyzing own emotions predicts lower levels of emotionality for the Dutch sample (Rieffe et al., 2008).
Understanding what triggered an emotion provides a sense of control, which could help children and adolescents to cope adequately with the situation instead of ruminating about it.
Moreover, for both the Dutch and Moroccan group understanding and attending to others’ emotions uniquely contributes to the prediction of social anxiety, but in reverse directions. For Dutch adolescents, attending to others’ emotions is related to fewer social anxious symptoms, whereas for their Moroccan peers this awareness is associated with more social anxiety. These differences may be due to cultural variations in the meaning and significance of emotions in daily lives. In cultures where emotions are often the subject of discourse, such as in the Dutch, attending to others’ emotions may promote interpersonal interactions. However, in cultures where emotions play a less dominant role in social exchanges and are less openly expressed, only socially anxious people may pay attention to and try to understand others’ emotional experiences. When filling out of the questionnaires, Moroccan adolescents often reported that they enjoyed doing this, but never thought about their emotions in such detail. Adolescents in the Netherlands did not make these kinds of remarks.
To conclude, the outcomes of this study indicate that adolescents’ social functioning, particularly having peer problems, and their understanding of their own and others’ emotions, both uniquely contribute to the prediction of internalizing problems.
Although adolescents’ social functioning and emotion awareness appear to be two relatively distinct constructs in the prediction of internalizing problems, together they cumulatively contribute to the prediction of social anxiety and emotionality.
Note, however, that these results are based on cross- sectional data, which prevents us from drawing conclusions concerning causal relations between the variables. Future longitudinal research should shed more light on the causal links that have been discussed in this article, further enhancing our
understanding of the relationship between mental health and emotional functioning in distinct cultural groups. In addition, this study primarily relied on self-reported questionnaires. Although this method is considered adequate to tap individuals’ private emotional states (DiBartolo & Grills, 2006), assessing youngsters’
social functioning in terms of behavior might benefit from including other informants, such as parents or teachers. A final recommendation for future research to broaden our understanding of adolescents’ internalizing problems is to incorporate other social skills, such as empathy, and other indices of emotional functioning, such as emotion regulation strategies.
The results would add to our findings indicating that various aspects of adolescents’ emotion awareness are uniquely associated with internalizing symptoms, over and above social functioning.
