Adolescents׳ perceptions of peers with depression: An attributional analysis
Introduction
Adolescent depression is receiving increasing attention in the psychological literature, which is unsurprising considering the prevalence rates reported in many studies. Lewinsohn et al. (1998) estimated that approximately 28% of adolescents will have experienced an episode of Major Depressive Disorder by the age of 19. Ireland׳s first comprehensive national study of youth mental health indicates that 30% of Irish adolescents score outside the normal range for depression (Dooley and Fitzgerald, 2012). Both studies indicate higher rates of depression for female adolescents.
However, very few studies have investigated the factors that predict the social acceptance or exclusion of adolescents with depression by their peer group. Social exclusion is linked to discrimination – a key aspect of the stigmatization process (Link and Phelan, 2001). This lack of focus is surprising given that research indicates that the peer group can offer some protection from depressive affect in adolescence (La Greca and Moore Harrison, 2005, Steinhausen and Metzke, 2001). Social exclusion from the peer group (i.e. peer exclusion) can take the form of exclusion from activities, being ignored, or receiving a lack of attention from peers. This type of peer exclusion is a significant predictor of emotional distress in adolescence (Kenny et al., 2013). Additionally, adolescents experiencing symptoms of depression experience decreased acceptance and support from their peers (Stice and Ragan, 2004) and the reasoning behind this is currently poorly understood.
To understand the factors that predict peer acceptance or exclusion, a strong theoretical model must be employed. However, research in this area is lacking in theoretically driven studies. One potentially powerful model which could be employed to address this gap is Attribution Theory (AT; Weiner, 1985), a well established theory in social psychology.
Attribution theory is a theory of social cognition which has been utilized to explain responses to people with mental disorders (Corrigan et al., 2003, Martin et al., 2000, Weiner et al., 1988). In essence, AT links the underlying structure of our causal thinking to the dynamics of emotions and actions (Weiner, 1985).
In its simplest form, AT hypothesizes that people make causal attributions about the behavior of others, these causal attributions give rise to emotions, and these emotions guide reactive behavior (Weiner, 1985, Weiner, 1986). A simple example is as follows. If somebody crashes into your car and you learn that they were texting on their mobile phone, you are likely to perceive the cause of the crash as controllable, you feel angry, and you react to the other person with annoyance. On the other hand, if you learn that the person who crashed into your car had a heart attack and lost control of their vehicle, you are more likely to perceive the cause of the crash as uncontrollable, and you experience sympathy which is likely to result in you providing assistance.
Weiner, 1985, Weiner, 1986 proposed various dimensions of causal attribution (locus, controllability, stability), but Anderson and Arnoult (1985) argued that ‘personal control’ (whether a cause is perceived by the attributor to be under volitional control) is the most important causal dimension. Weiner (1988) highlighted that the key aspect of the attribution model is the mediating role of emotions as determinants of an individual׳s subsequent behavior (e.g. social acceptance/exclusion). In particular, emotions of anger and pity are thought to influence behavior towards others. Research has provided evidence that the perception of controllability is the main causal dimension linked with these emotions (Meyer and Mulherin, 1980, Weiner et al., 1982). Anger is greater in the case of controllable causes, whereas pity is greater when the cause of behavior is perceived as uncontrollable (Weiner et al., 1982).
Weiner (1993) argued that when reacting to stigmatized individuals, observers/attributors search for causation, and also judge personal responsibility. Whilst causal controllability is a major determinant of responsibility, they differ in that controllability is a causal property, whereas responsibility is an inference or judgment about a person (Weiner, 1993, Weiner, 1995). Sometimes cause is located within the person and is controllable, but still responsibility is not inferred e.g. a business failure due to caring for a sick relative (Weiner, 1995). Therefore, in incorporating responsibility into AT, Weiner (1993) proposed a sequence whereby personal control predicts inferences of responsibility, which influences emotional reactions, and these emotional reactions predict behavior.
Juvonen (1991) showed that sixth grade students rate certain types of “deviant” peers more negatively than others, depending on the level of responsibility attributed to the behavior. For example, children perceived bragging and aggressive peers as more responsible for their behavior and thus they evoked more negative emotions (anger and irritation) than peers who were socially withdrawn, or had a physical condition. In line with AT, Juvonen concluded that the more a peer induced positive emotions (sympathy and pity), the more likely children were to support them. Conversely, the more a peer elicited negative emotions (anger and irritation), the more likely they were to be excluded. In terms of depression, there is evidence that children rate a depressed peer with high life stress more positively than a depressed peer with low life stress (Peterson et al., 1985). In other words, attributing depression to external stress reduces the tendency to act negatively towards a depressed peer. If attributions of personal control over the cause of a disorder can influence subsequent behavior, this may have implications for the design and implementation of anti-stigma interventions.
Reported rates of depression increase more steeply for girls than boys during adolescence. Nolen-Hoeksema and Girgus (1994) claim that after the age of 15, girls are about twice as likely to be depressed as boys, indicating that mid-late adolescence is a key time to investigate gender differences in adolescent responses to peers with depression.
Research has also highlighted that adolescents have differential attitudes towards male peers with mental disorders and female peers with mental disorders. In light of AT, recent research indicates that greater belief in a child׳s responsibility for his condition (ADHD or depression) is associated with lower acceptance for males only (Swords et al., 2011). The authors suggest that further work is necessary to investigate why attributions of responsibility have a greater influence in judgments of boys in comparison to girls. Although participants in the Swords et al. (2011) study responded to both male and female hypothetical peers, the majority of studies examining perceptions of aggression and withdrawal have used only males as the stimuli (e.g. Juvonen, 1991, Peterson et al., 1985, Sigelman and Begley, 1987). This has left gaps in the literature in terms of our understanding of the effect the vignette character׳s gender has on responses.
We employ a strong theoretical framework by applying a fuller model of AT than has been previously used, to test adolescents׳ causal attributions of personal control, inferences of responsibility, emotional reactions, and peer acceptance of a hypothetical peer presented as clinically depressed. Our study builds on previous research in several ways.
Much of the mental health literature is based on adolescents׳ understanding of mental disorders as general concepts rather than how they understand specific symptoms (e.g. feeling anxious) which may represent preliminary manifestations of more serious mental health difficulties (MacLean et al., 2012). Therefore the vignette in this study focuses on specific symptoms of depression without explicitly labeling it.
Unlike previous research that has explored this issue from an attributional perspective, this study doesn't provide participants with information regarding the cause of the depressed behavior; rather, they will be asked to suggest a possible cause of such behavior. This method of using a free-response procedure was deemed as an appropriate research procedure by Weiner (1985). In addition, McAuley et al. (1992) criticized research in the area of AT for translating participants׳ causal attributions into the causal dimensions and therefore committing the fundamental attribution research error. In other words, previous research assumed that the researcher and participant perceived causes in the same way (for example, assuming that all people view genetic causes as low on personal control and environmental causes as higher on personal control). To avoid this, this study asks participants to explicitly indicate the level of personal control they attribute to their suggested cause of depression.
Consistent with previous attribution research (Juvonen, 1991, Weiner, 1993, Weiner et al., 1988), our study hypothesizes that causes of depressive behavior perceived as uncontrollable will not elicit inferences of responsibility, pity/sympathy will be elicited, and this will result in peer acceptance. Conversely, when the cause of depressed behavior is perceived as controllable, inferences of responsibility and the emotion of anger/irritation will be elicited, which will result in peer exclusion.
Section snippets
Participants
Participants were 401 adolescents with an age range of 14.75–17.08 years (M=15.90 years; S.D.=0.403 years). The gender distribution of the sample was 53.1% female (n=213) to 46.9% (n=188) male. Adolescents were recruited from 4th year (students in their 12th year of school). These adolescents were recruited from ten, non fee paying, single sex secondary schools, selected from schools with 600+ students on the Irish Department of Education published lists of schools in South-East Ireland.
Results
Structural equation modeling was used to model the relationships between causal attributions of personal control, responsibility, positive emotions, negative emotions, and peer acceptance. All statistical analyses were conducted using AMOS version 18.
Summary findings
This study employed Structural Equation Modeling to test the pathways predicted by Attribution Theory (AT; Weiner, 1985, Weiner, 1986, Weiner, 1993, Weiner, 1995) separately for male and female adolescents with depression. Our findings suggest that peer acceptance of both male and female adolescents with depression is predicted by pathways whereby cause of depression is perceived as uncontrollable, responsibility is not inferred, and sympathy/pity is felt by the participant. These findings are
Conclusion
This study furthers our understanding of how teenagers evaluate and react to the behavior of a peer with depression. Key findings indicate that peer group acceptance of an adolescent with depression is predicted by low attributions of personal control, low inferences of responsibility, and emotional reactions of sympathy/pity rather than anger/irritation. When responding to a male peer with depression, male participants perceive them as having more personal control over the causes of their
Acknowledgments
We thank the adolescents who participated in this study, their parents who gave permission for their participation and the schools who provided the time and space to facilitate the collection of data.
References (41)
- et al.
Interpersonal relationships and emotional distress in adolescence
Journal of Adolescence
(2013) - et al.
Biogenetic explanations and stigma: a meta-analytic review of associations among laypeople
Social Science & Medicine
(2013) - et al.
Major depressive disorder in older adolescents: prevalence, risk factors and clinical implications
Clinical Psychology Review
(1998) - et al.
Psychiatric labels and other influences on young people׳s stigmatizing attitudes: findings from an Australian national survey
Journal of Affective Disorders
(2013) Integrative Guide for the 1991 CBCL/4-18, YSR, and TRF Profiles
(1991)Diagnostic and Statistical Manual of Mental Disorders
(1994)- et al.
Attributional style and the everyday problems in living: depression, loneliness, and shyness
Social Cognition
(1985) - et al.
Testing Structural Equation Models
(1993) - et al.
Alternative ways of assessing model fit
- et al.
An attribution model of public discrimination towards persons with mental illness
Journal of Health and Social Behavior
(2003)
At issue: stop the stigma: call mental illness a brain disease
Schizophrenia Bulletin
My World Survey. National Study of Youth Mental Health
Can brief workshop interventions change care staff understanding of challenging behaviours?
Journal of Applied Research in Intellectual Disabilities
What do implicit measures tell us? Scrutinizing the validity of three common assumptions
Perspectives on Psychological Science
Development of depression from preadolescence to young adulthood: emerging gender differences in a 10-year longitudinal study
Journal of Abnormal Psychology
Stigma as related to mental disorders
Annual Review of Clinical Psychology
Staff reactions to self-injurious behaviours in learning disability services: attributions, emotional responses and helping
British Journal of Clinical Psychology
Deviance, perceived responsibility, and negative peer reactions
Developmental Psychology
Adolescent peer relations, friendships, and romantic relationships: do they predict social anxiety and depression?
Journal of Clinical Child and Adolescent Psychiatry
Conceptualizing stigma
Annual Review of Sociology
Cited by (19)
Understanding stigma toward schizophrenia
2022, Psychiatry ResearchSocial Anxiety and Depression Stigma Among Adolescents
2021, Journal of Affective DisordersCitation Excerpt :Each participant read two vignettes, one female and one male, one with social anxiety disorder and one with depression. Both vignettes were validated by clinicians to ensure they met criteria for clinical diagnosis (see Dolphin and Hennessy, 2014, for validation procedure for depression vignette, this study employed the same process to validate the social anxiety disorder vignette). Four versions of the questionnaires were distributed in each school to ensure gender balance and to balance the order of presentation of the vignettes: a) social anxiety disorder (F) + depression (M); b) depression (M) + social anxiety (F); c) social anxiety disorder (M) + depression (F); d) depression (F) + social anxiety (M).
Reasons for disliking a classmate: A comparative study with Spanish and Portuguese students
2021, International Journal of Educational ResearchCitation Excerpt :Furthermore, any student in a class, and not only those students displaying problematic behaviour or specific conditions, might be disliked by his/her classmates. Thus, these studies lack ecological validity (Dolphin and Hennessy, 2014). In order to overcome these limitations, the current study aims to examine students’ reasons for disliking a classmate within the context of regular classroom life and processes.
A systematic review of depression literacy: Knowledge, help-seeking and stigmatising attitudes among adolescents
2019, Journal of AdolescenceCitation Excerpt :The study characteristics and the various depression literacy components reported in the 50 studies are shown in Table 1. Vignette-based methodology was the most commonly used method of assessing depression literacy, used in by 29 (58%) studies (Adeosun, 2016; Aluh et al., 2018; Ando et al., 2018; Attygalle et al., 2017; Bruno et al., 2015; Burns & Rapee, 2006; Coles et al., 2016; Dolphin & Hennessy, 2014; Essau et al., 2013; Hernan, Philpot, Edmonds, & Reddy, 2010; Jorm et al., 2006; Kelly & Jorm, 2007; Kelly et al., 2006; Lam, 2014; Lee et al., 2014; Lubman et al., 2017; Marcell & Halpern-Felsher, 2007; Marshall & Dunstan, 2011; Mason et al., 2015; McCarthy et al., 2011; Melas et al., 2013; Ogorchukwu et al., 2016; Olivari & Guzmán-González, 2017; Olsson & Kennedy, 2010; O’Driscoll et al., 2012; Sawyer et al., 2012; Sharma et al., 2017; Swords, Hennessy, & Heary, 2011; Yoshioka, Reavley, MacKinnon, & Jorm, 2014). The vignettes used varied across studies in terms of gender (the majority of studies (62%) used vignettes of a single gender) and the symptoms featured in the vignettes.
Social contact as a strategy for self-stigma reduction in young adults and adolescents with mental health problems
2018, Psychiatry ResearchCitation Excerpt :As demonstrated in other investigations (Corrigan et al., 2005), other factors must be taken into account when studying familiarity, such as the nature or type of contact that, in some cases, may even be associated with significant elevations of stigma. It is possible that the low levels of public stigma in our study: a) are associated with the willingness to participate in this type of social inclusion programs in which all participants know that they will interact with people with and without mental health problems; and, b) support that our data does not go "apparently" in the line of studies showing that familiarity correlates with a lower level of stigma in the adult (Angermeyer and Matschinger, 1996) and adolescent population (Dolphin and Hennessy, 2014; Martínez-Hidalgo, 2015; Swords et al., 2011; Yap et al., 2013). Although some studies that support the existence of a significant relationship between public stigma and self-stigma have been found in the literature (Muñoz et al., 2011a, 2011b), the data obtained in this work is in line with those found by other authors (Corrigan et al., 2006) who have not found a significant relationship between both variables.
The stigma of mental illness in children and adolescents: A systematic review
2016, Psychiatry ResearchCitation Excerpt :The degree of blame varied by diagnosis (e.g. Corrigan et al., 2005). High caliber qualitative and quantitative means found that where causal attributions were identified as beyond the child's control, YP-MHD were not blamed (Peterson et al.,1985; Dolphin and Hennessy, 2014; O’Driscoll et al., 2015a, 2015b). Eighteen studies, of varying scientific rigor, examined social distance/behavioral intentions, making it the most frequently assessed construct of stigmatization (Table 1).