Journal of Affective Disorders xxx (xxxx) xxx

Contents lists available at ScienceDirect

Journal of Affective Disorders

journal homepage: www.elsevier.com/locate/jad

Research paper

Social Anxiety and Depression Stigma Among Adolescents

*
Hannah Lynch, Cal McDonagh, Eilis Hennessy
University College Dublin

ABSTRACT

Background: The stigma associated with mental disorders in adolescence has a range of detrimental consequences, negatively impacting help-seeking behaviours and
quality of life. Social anxiety typically has its onset during adolescence, but the associated stigma is not well understood. This study was designed to improve
understanding of social anxiety and depression stigma in adolescence by examining demographic and personal variables that predict them both.
Methods: Three hundred and fifteen adolescents (150 males, 165 females), from 2nd and 4th year of secondary school (mean ages 13.92 and 15.92 years
respectively), completed the Peer Mental Health Stigmatization Scale to measure both personal and perceived stigma towards vignettes depicting social anxiety and
depression. They also answered demographic questions and completed the Social Anxiety Scale for Adolescents.
Results: Hierarchical multiple regression revealed that gender predicted both personal and perceived stigma towards depression and personal stigma towards social
anxiety. Males scored higher on measures of personal stigma whereas females had higher scores on perceived stigma. Ability to accurately identify symptoms
described in a vignette, predicted lower personal stigma towards social anxiety and higher perceived stigma towards depression. Depression was more stigmatized
than social anxiety.
Limitations: The study used vignettes to represent individuals with depression and social anxiety so participant responses may not reflect their behaviour towards real
peers.
Conclusions: Findings emphasise the importance of separately considering personal and perceived stigma for each condition and highlight gender differences in stigma
responses that need to be considered in intervention design.

1. Introduction Kogan et al., 2016; World Health Organization, 2018) or social phobia.
The symptoms of social anxiety disorder typically appear during
Adolescence is the peak time for the onset of a wide range of mental adolescence (Keller, 2003), with reported lifetime prevalence of 9%
disorders including depression and social anxiety disorder (Belfer, 2008). among US 13-to-18-year-olds (Merikangas et al., 2010) and 12%
While the challenges faced by young people diagnosed with these among Dutch 19-year-olds (Ormel et al., 2015). The same studies
disorders are well documented and significant research is devoted to report higher prevalence of depression: lifetime prevalence of 12%
developing and evaluating therapeutic interventions, much less research among US 13-to-18-year-olds (Merikangas et al., 2010) and 15%
has focused on the social consequences of these diagnoses and in among Dutch 19-year-olds (Ormel et al., 2015).
particular on the associated stigma. Yet the importance of understanding The present article reports on the findings of a study that adds to
mental health stigma is highlighted by research demonstrating negative the relatively small body of research on stigma associated with social
consequences for adolescents who have experienced mental disorders anx-iety disorder. In order to facilitate interpretation of the findings, the
including social exclusion, hurtful treatment and reduced expectations study also gathered data on depression stigma, which has been
(McKeague et al., 2015a; Moses 2014). studied much more extensively in adolescence. The findings have
According to the World Health Organisation (2017), anxiety disor-ders implications for the design of mental health literacy programmes and
and depression have similar global prevalence rates in adolescence. for our un-derstanding of the peer context of young people presenting
Despite this fact, research investigating stigma has paid much more with symptoms of depression or social anxiety disorder.
attention to adolescent depression than to social anxiety disorder. Although
social fears are common in adolescence, high levels of social anxiety, 1.1. Stigma
causing clinically significant distress and impairment, are described by the
two major diagnostic systems and typically referred to as social anxiety Stigma is a complex concept, that can be conceptualised as a com-
disorder (American Psychiatric Association, 2013;
bination of negative stereotypes, prejudices and discrimination in the

* Corresponding author at: School of Psychology, University College Dublin, Belfield, Dublin 4, Ireland.
E-mail address: eilis.hennessy@ucd.ie (E. Hennessy).

https://doi.org/10.1016/j.jad.2020.11.073
Received 30 March 2020; Received in revised form 6 June 2020; Accepted 8 November 2020
Available online 14 November 2020
0165-0327/© 2020 The Author(s). Published by Elsevier B.V. This is an open access article under the CC BY license (http://creativecommons.org/licenses/by/4.0/).

Please cite this article as: Hannah Lynch, Journal of Affective Disorders, https://doi.org/10.1016/j.jad.2020.11.073
H. Lynch et al. context of a power differential (Corrigan and Shapiro, 2010). The
complexity of the concept has also resulted in the further specification of
different forms of stigma (Heary et al., 2017). These include: personal Journal of Affective Disorders xxx (xxxx) xxx
stigma, which refers to one’s own cognitive, emotional and behavioural
responses towards an individual with a mental health condition; perceived Their study included a vignette depicting social phobia, however, these
(public) stigma, which refers to one’s perception of others’ re sponses data were combined with data from vignettes depicting a range of
towards an individual with a diagnosed mental health disorder; and self- other mental disorders meaning that it is not possible to chart the
stigma, which refers to the internalization of perceived stigma by a person changes associated with any single disorder. Calear et al. (2011), in
who has a diagnosed mental health disorder (i.e. cognitive or emotional contrast focused only on the stigma associated with depression and
responses towards the self); personal stigma and perceived stigma are found that perceived stigma was higher among older adolescents
relevant to the present paper. whereas personal stigma was higher among younger groups. The
Personal and perceived stigma are thus conceptually distinct and authors explain this finding in relation to the older adolescents having
they are associated with different outcomes. For example, Nearchou et more opportunity to learn about nature of depression.
al. (2018) found that perceived, but not personal, stigma predicted
adolescents’ help-seeking intentions for both anxiety and self harm. 1.4. Recognition of social anxiety disorder
Calear et al. (2017) found that personal and perceived stigma towards
individuals with generalised anxiety disorder were differently predicted There is some evidence to suggest that adolescents who recognise and
by adolescents’ family composition, mental health literacy and scores accurately label the symptoms associated with common psychiatric
on a measure of depression symptoms. Mental health literacy is diagnoses (e.g. depression) have lower scores on stigma scales, possibly
defined as ‘understanding how to obtain and maintain positive mental reflecting a higher level of mental health literacy. For example, Yap et al.
health; understanding mental disorders and their treatments; (2013) presented adolesents with vignettes describing a young person with
decreasing stigma related to mental disorders; and enhancing help- symptoms of a range of common psychiatric diagnoses. Across all five
seeking efficacy’ (Kutcher et al., 2016 p. 155). vignettes, respondents who used accurate psychiatric labels (including
Research on adolescent mental health stigma is particularly impor tant depression and social phobia) were more likely to describe a young person
because adolescence is a time when the influence and importance of peer in a vignette as ‘sick’ rather than ‘weak’.
relationships and opinions substantially increases (Deater-Deak ard, 2001),
making adolescents more sensitive to stigma. Problems with peer 1.5. Participant social anxiety symptoms
relationships have been associated with psychopathology including anxiety
and social anxiety disorders (Masten et al., 2009; Festa and Ginsburg, It might be assumed that having personal experience of mental distress
2011). Moreover the peak age of onset for mental disorders is early or experiencing symptoms of a mental disorder would be asso ciated with
adolescence, as cognitive and emotional changes are occurring at this time lower levels of personal stigma, but this is not always the case. For
(Hertzman and Boyce, 2010) and there is evidence to suggest that example, Dardas et al. (2017) found no association between participant
adolescents express a range of negative responses towards de scriptions scores on a measure of depression and either their personal stigma score
of anxious peers including a belief that they are ‘weak’, that they are or their perceived stigma score. Similarly, Busby Grant et al (2016) found
personally responsible for their symptoms and are deserving of blame that among university students, depression symptoms did not predict
(Hanlon and Swords, 2019). personal depression stigma. However, in a recent study Calear et al.,
Identifying variables which affect stigma, at a time when young (2017) found that depression symptoms predicted perceived anxiety
people are most sensitive to stigma and when mental disorders are stigma. In the present study we decided to include a measure of individual
most likely to develop, can identify issues that need to be targeted and social anxiety symptoms because it could be argued that individuals who
help to reduce stigma in a timely and effective manner. In the absence are more socially anxious might also be more alert to perceived stigma of
of in formation on the personal and socio-demographic variables both depression and social anxiety disorder, because of their inherent fear
associated with stigma of social anxiety disorder, we explored the of negative evaluation. Furthermore, personal stigma of depression and
variables that have been found to predict the stigma of other mental social anxiety might be heightened among this group due to greater
disorders in adolescence. preference for social dis tance. However because so little is understood
about the association between social anxiety and the stigma of mental
1.2. Gender
disorders in adoles cence, we did not include a specific hypothesis to this
effect.
Several studies on depression stigma in adolescence have found that
adolescent males have higher scores on personal stigma of depression 1.6. The present study
(Calear et al., 2011; Griffiths et al., 2008). The stigma associated with
anxiety disorders has been less thoroughly researched (Curcio and This study was designed to fill the gaps identified in the literature
Corboy, 2019). However, Busby Grant et al. (2016) and Calear et al. (2017) above, with three primary aims: first, to clarify the predictors of
report that male gender was a significant predictor of higher personal personal and perceived social anxiety disorder stigma among
stigma towards anxiety disorders and Calear et al. (2017) also found that adolescents; sec ond, to clarify the predictors of personal and
perceived depression stigma among adolescents; and third, to
male gender predicted perceived stigma. Both of these studies used the
compare social anxiety disorder stigma levels to depression stigma.
same measure of stigma towards generalised anxiety disorder and to date
Previous studies which have explored the predictors of mental disorder
we are not aware of any research on stigma associated with social anxiety
stigma in adolescence, have focused predominantly on depression.
disorder among adolescents.
Therefore, depression acts as a useful comparator for under-
1.3. Age researched conditions such as social anxiety.
Specifically, this study investigated the effects of gender, age,
Several previous studies about adolescent mental health stigma recognition of social anxiety disorder and depression and current social
have suggested that there are age related changes, however these anxiety symptoms on personal and perceived social anxiety disorder and
are not always easy to interpret. For example, Jorm and Wright (2008) depression stigma. Both personal and perceived stigma were measured
found that desire for social distance from a peer with a mental disorder because previous research has indicated that they may have different
pedictors. A better understanding of the nature of social anxiety disorder
decreased with age, however perception of others’ stigma increased.
and depression stigma in adolescence has potential to provide insight into
the peer reactions that may be experienced by those who have been
diagnosed with these conditions and to provide useful information to those
who wish to design educational interventions to increase mental

2
H. Lynch et al. health literacy e.g. by identifying a need to target interventions to spe-
cific age groups or those with diagnosed mental disorders.
2. Method Journal of Affective Disorders xxx (xxxx) xxx

2.1. Participants Table 2

Descriptive statistics for Peer Mental Health Stigmatization Scale (PMHSS) and
Three hundred and fifteen secondary school students completed the Social Anxiety Scale for Adolescents (SAS-A)
the questionnaire (150 males, 165 females), across 7 schools in the
Variable N Range Cronbach Mean Skewness/
eastern region of the Republic of Ireland. These schools were situated
α (SD) kurtosis
in rural and urban regions, and one large town. The age range of
students was between 13 and 17 years. In Ireland the education Peer Mental Health Stigmatization Scale (PMHSS) sub-scales
system comprises 8 years in primary school and a further 6 years in Perceived Social 309 8-40 .81 22.48 .02/-.34
secondary school. Stu-dents typically enter secondary school at age Anxiety stigma (5.48)
nd th
12/13. Participants included in this study were 2 year and 4 year Female 162 23.18
secondary school students. The schools were chosen through (5.89)
convenience sampling. Data was collected between January and May Male 147 21.71
2018. Table 1 depicts the de-mographic characteristics of the sample.
(4.89)
2.2. Ethics Personal Social 307 8-40 .72 16.12 .38/.04
Anxiety stigma (4.26)
Ethics approval was granted by the University College Dublin Female 164 15.27

(UCD) Human Research Ethics Committee - Humanities. (4.03)

Male 143 17.10
2.3. Procedure
(4.32)
Perceived 309 8-40 .79 24.35 -.32/.14
Twenty schools were contacted via email where a brief description
Depression (5.29)
of the study was given with information sheets for the school principal.
Emails were followed up with phone calls to determine if the school stigma

was interested in taking part. If a school was willing to participate, re- Female 163 25.21

searchers visited the schools with information sheets for parents and (5.15)
students. A parental consent form indicating consent for their child to Male 146 23.38
be invited to take part in the study was also given. Before handing out (5.29)
the questionnaires, written participant assent was obtained. Personal Depression 307 8-40 .75 17.48 .24/.34
stigma (4.67)
2.4. Measures
Female 162 16.84

The Social Anxiety Scale for Adolescents (SAS-A) (Storch et al., (4.65)
2004): A 22 item measure of social anxiety symptoms for use with Male 145 18.19
ages 7 to 19 years. Each statement is rated on a 5 point likert scale (5.60)
(not at all, hardly ever, sometimes, most of the time, all of the time).
Social Anxiety Scale for Adolescents (SAS-A)
Higher scores reflect greater social anxiety symptoms. The total score
SAS-A total 307 18-90 .86 45.88 .41/.40
of the SAS-A demonstrated acceptable internal reliability in the present
study, de-tails are in Table 2. (11.82)
Clinical Vignettes: The vignettes (each 260 words), depicted a character Female 162 49.87
displaying symptoms of depression or social anxiety disorder who was (11.68)
described as ‘the same age’ as the participants. Each participant read two Male 145 41.42
vignettes, one female and one male, one with social anxiety disorder and (10.31)
one with depression. Both vignettes were validated by clinicians to ensure
they met criteria for clinical diagnosis (see Dolphin and Hennessy, 2014, for
(M) + depression (F); d) depression (F) + social anxiety (M).
validation procedure for depression vignette, this study employed the same
Ability to correctly identify the disorder presented in the vignette
process to validate the social anxiety disorder vignette). Four versions of
was assessed through an open-ended question which asked “What, if
the questionnaires were distributed in each school to ensure gender anything do you think is wrong with Neil/Kate?” Participants’ responses
balance and to balance the order of presentation of the vignettes: a) social were coded 1 if they could identify the disorder and 0 if they could not.
anxiety disorder (F) + depression (M); b) depression (M) + social anxiety Accurate depression responses included those that mentioned
(F); c) social anxiety disorder “depres-sion” or “depressed” (Dolphin and Hennessy, 2016). Accurate
social anxiety responses included those that used one of the following
Table 1
“social anxiety” or “socially anxious”. Words such as “perhaps” or
Demographic characteristics of sample.
“probably has” were ignored. Given the low rates of those who used an
Year in School Mean Age (yrs) SD exact social anxiety definition the definition was expanded to include
nd “anxiety” and “anxious” (Yap et al., 2013).
2 Year (N=167) 13.91 .44
th
Peer Mental Health Stigmatization Scale (PMHSS) (McKeague et
4 Year (N=148) 15.92 .57
nd th
al., 2015b): This scale was developed for use with adolescents and
Gender Frequency 2 Year 4 Year includes two subscales: perceived stigma (8 items); personal stigma (8
Male 150 (47.6%) 90 88 items), with a further 8 positively worded items that do not form part of
Female 165 (52.4%) 77 60 the scale. Responses are measured using a 5 point likert scale
(disagree completely, disagree, neither agree nor disagree, agree,
agree completely). Personal stigma assessed the participants’ own
responses to a hypothetical peer (e.g., “I believe that teenagers like
Neil are dangerous”) whilst perceived stigma assessed their
perception of “most people’s” responses a peer (e.g., “Most people
believe that teenagers like Neil are to blame for their problems”). Total
scores were calculated by adding the participant’s item scores, with
higher scores indicative of greater personal, perceived or overall
stigma. Cronbach’s alphas in the present study demonstrated internal consistency was also found for personal stigma using the
acceptable internal consistency for the perceived stigma scale for depression (α=.75) and social anxiety disorder vi-gnettes (α=.72).
depression (α=.79) and social anxiety disor-der (α=.81). Acceptable
3
H. Lynch et al. Journal of Affective Disorders xxx (xxxx) xxx

2.5. Data analysis 3.4. Predictors of social anxiety disorder stigma

Hierarchical multiple regression analyses were used to explore the Personal stigma towards social anxiety disorder was considered first.
first and second research aims, determining the effect of each Gender (0 = female, 1 = male) and age (in years) were entered at the first
indepen-dent variable (age, gender, identification and SAS-A score) on stage, this model was statistically significant, F (2,296) = 8.45, p < 0.05,
explaining 5.4% of variance in the dependent variable. Gender made a
both stigma subscales (personal and perceived stigma) for both
significant contribution to the model, with males reporting higher levels of
disorders. To explore the third research aim, comparing depression
personal stigma than females. Age was not a significant contributor. During
and social anxiety disorder stigma, dependent t-tests were used.
the second stage, identification of social anxiety disorder and SAS-A score
Analyses were all con-ducted using SPSS v24. were added to the model. Variance explained rose to 8.5%, this was a
2
3. Results statistically significant increase (R = .085; F (2,294) = 5.01, p < 0.05).
Identification of social anxiety disorder contributed significantly to the
3.1. Missing data model, accuracy was associated with lower stigma; however SAS-A score
did not contribute significantly. The final model was significant F (4, 294) =
Less than 5% of data were missing for any variable, thus it was 6.84, p < 0.05.
decided that the missing data could be ignored without any significant Pearson correlations demonstrated that none of the predictor vari-
ramifications on the analyses (Schafer, 1999). Pair wise deletion was ables were significantly correlated with the dependent variable of
used for participants with missing data. perceived social anxiety disorder stigma. As such it was unsuitable to
conduct hierarchical regression models for perceived stigma.
3.2. Descriptive statistics
3.5. Predictors of depression stigma
Descriptive statistics for each of the dependent variables can be seen in
Another hierarchical multiple regression was performed to investi-
Table 2. Social anxiety disorder was accurately identified by 47% of
gate the predictors of personal depression stigma. In the first step, age
participants. Depression was accurately identified by 67% of partici-pants.
and gender were entered. This model was statistically significant F(2,
Chi square tests were used to explore gender and year in school with
299) = 5.37; p < .05 and explained 3.5 % of variance in personal
ability to accurately identify depression and social anxiety. Iden-tification of
stigma. Both variables made a significant contribution to the model;
social anxiety disorder and gender were not independent
2
being male and younger were associated with greater stigma. During
χ (1, N = 315) = 23.57, p < 0.05. This was also the case for depression the second stage, identification of depression and SAS-A score were
2 added to the model. The total variance explained by the model was
χ (1, N = 315) = 19.65, p < 0.05. In both cases significantly more fe-males
correctly identified the disorder than expected. Identification of social 4.5%. This model was significant F(4, 299) = 3.44; p < .05. Neither
2
anxiety disorder and year group were not independent, χ (1, N = 315) =
correct identification of depression nor SAS-A scores were significant
13.87, p < 0.05. The fourth year group were significantly more likely to predictors of personal stigma.
accurately identify social anxiety than expected, but there were no A final regression model was conducted to investigate the
2 predictors of perceived depression stigma. Gender and age were
significant differences for the identification of depression χ (1, N = 315) =
entered at stage one of the model. This model was significant F (2,
1.98, p > 0.05. These results suggest that ability to identify social anxiety
disorder is associated with being older and female. Whereas ability to
300) = 4.66; p < .05, explaining 3% of variance. Gender contributed
identify depression is only associated with female gender and is significantly to the model whilst age did not. Being female was
independent of year group. associated with higher perceived stigma. At stage two of the model the
variables correct iden-tification of depression and SAS-A score were
3.3. Hierarchical multiple regression entered. This model was also significant F (4, 300) = 5.75; p < .05.
2
Variance explained rose to 7.2%, this was a significant increase (R
Hierarchical multiple regression (HMR) models were conducted to Change = .042; F(2, 296) = 6.7; p < .05). Accurate identification of
explore the predictors of personal and perceived social anxiety depression was associated with higher perceived stigma scores, whilst
disorder and depression stigma. Tables 3 and 4 present the results of SAS-A score did not contribute significantly to the model.
these ana-lyses by stigma type. Prior to conducting the analyses, the
assumptions of HMR were tested. The sample size of 315 students 3.6. Stigma comparison
was an adequate size to conduct HMR given the four independent
To explore the third research question, comparing social anxiety
variables (Tabachnick and Fidell, 2001). Assumptions of normality,
linearity and homosce-dasticity were satisfied via scatter plots and disorder and depression stigma, paired sample t-tests were used.
residual plots (Pallant, 2001). Perceived stigma toward depression (M = 24.28, SD = 5.25) was

Table 3
Hierarchical multiple linear regression models for personal social anxiety stigma.

2 2
Stigma Type Step Predictor B SE β t R R change

Personal Stigma
1 5.4*
Gender 1.80 .48 .21 3.73*
Age -.33 .22 -.08 -1.54
2 8.5* 3.1*
Gender 1.12 .53 .13 2.14*
Age -.20 .22 -.05 -.94
Identification -1.23 .50 -.15 -2.45*
SAS-A score -.04 .02 -.12 -1.93
2
* = p < 0.05; B = Unstandardized coefficient; SE = Standard Error; β = Standardized coefficient; t = t statistic; R = Coefficient of determination; Identification = correct
identification of social anxiety disorder vignette
4
H. Lynch et al. Journal of Affective Disorders xxx (xxxx) xxx

Table 4
Hierarchical multiple linear regression models for personal and perceived depression stigma.

2 2
Stigma Type Step Predictor B SE β t R R change

Personal Stigma
1 3.5*
Gender 1.29 .53 .14 2.43*
Age -.49 .24 -.12 -2.08*
2 4.5 1
Gender .89 .59 .096 1.52
Age -.47 .24 -.11 -1.96
Identification -.83 .58 -.08 -1.42
SAS-A score -.03 .02 -.06 -1.03

Perceived Stigma
1 3*
Gender -1.83 .60 -.17 -3.03*
Age .06 .27 .012 .22
2 7.2* 4.2*
Gender -.95 .65 -.09 -1.45
Age -.002 .27 .00 -.01
Identification 2.13 .65 .19 3.28*
SAS-A score .046 .027 .10 1.72
2
* = p < 0.05; B = Unstandardized coefficient; SE = Standard Error; β = Standardized coefficient; t = t statistic; R = Coefficient of determination; Identification = correct
identification of depression vignette

significantly greater than towards social anxiety disorder (M = 22.50, predict stigma generally across conditions. Variables which predict
SD = 5.50); t(302) = -6.23 p < 0.05, d = 0.36. A paired samples t-test perceived stigma in other conditions may not be
was also conducted to compare personal stigma towards depression
and social anxiety disorder. Again, personal stigma toward depression
(M = 17.44, SD = 4.65) was significantly greater than toward social
anxiety disorder (M = 16.15, SD = 4.20); t(298) = -6.55, p < 0.05, d =
0.38. These results suggest that adolescents are distinguishing
between the two vignettes and are responding with more stigma
towards the depic-tion of the teenager with depression.
4. Discussion

4.1. Summary of findings and comparison to previous literature

This was the first study of its kind in which the predictors of
personal and perceived stigma towards social anxiety disorder and
depression were investigated in an adolescent sample. Findings
indicate that per-sonal and perceived stigma are predicted by different
variables. Furthermore this study found depression was more
stigmatised than social anxiety disorder among adolescents.
The first aim of this study was to explore the predictors of personal
and perceived stigma associated with social anxiety disorder. Greater
personal stigma was predicted by being male and by an inability to
identify social anxiety disorder. There was no effect of age or SAS-A
score. The gender findings are consistent with previous literature look-
ing at stigma of anxiety disorders more generally (Griffiths et al., 2008;
Jorm and Wright, 2008). Wright and Jorm (2009) suggested that
because there is a higher prevalence of anxiety among females,
adoles-cent girls have more exposure to symptoms of anxiety and this
may contribute to this finding. Our chi square analyses also
demonstrated that females were more likely to correctly identify both
social anxiety disorder and depression. Coles et al. (2016) also found
that girls were significantly more likely to correctly identify depression
but they found no such gender difference for social anxiety disorder.
The results for perceived social anxiety disorder stigma indicate that the
variables examined were not associated with stigma, demonstrated by the
lack of correlations between the predictor variables with the outcome
variable, thus eliminating the possibility of performing a regression analysis.
Whilst not adding to our understanding of variables directly predicting
perceived stigma, these findings allow appreciation of the varied nature of
stigma constructs. It reinforces the fact that the same variables may not
important factors for perceived social anxiety stigma. This also high- In relation to perceived depression stigma, higher score (belief that
lights the need for disorder-specific research. there is greater stigma in the general population) was predicted by fe-
The second aim of this study was to explore depression stigma. For male gender and correct identification of depression. Similar gender
personal depression stigma, being male and being younger were both differences were reported by Busby Grant et al. (2016) and Calear et
associated with higher stigma. These findings are consistent with pre-vious
al. (2011) in relation to depression. Although the findings don’t provide
research (Busby Grant et al., 2016; Griffiths et al., 2008). The model
us with an explanation for these gender differences, it is worth noting
that they are consistent within findings related to gender differences in
exploring predictors of personal depression stigma was the only one which
social cognition. For example, females have higher scores in empathy
showed a significant effect of age. This was surprising given the many
(Jolliffe and Farrington, 2006), mindreading (Wacker et al., 2017) and
cognitive, emotional and social changes that occur across adolescence.
social perspective taking (Flannery et al., 2017) suggesting that they
However, this may be attributable to the relatively narrow age range of
may be more aware of and focused on the perspective of others.
participants. Perhaps this was not large enough to allow age differences to
These gender differences in social cognition may also in part explain
be detected. Thus it would be beneficial for future research to explore
the findings that males had higher levels of personal stigma towards
stigma within a wider age range of adolescents.
both social anxiety and depression because they imply that, on
Finally, the finding that correct identification of depression did not
average, males would be less able to take the perspective of an
predict personal depression stigma contrasts with findings by Dolphin and
individual with a mental health disorder.
Hennessy (2016) who found that correct identification of depression was
We argued that it would be logical to expect that social anxiety
associated with lower scores on negative emotional reactions. However,
Dolphin and Hennessy (2016) only found significant differ-ences on the symptoms might predict perceived stigma because of the nature of social
emotional response of “anger”, such that young people who correctly anxiety. This was not the case. Neither personal or perceived stigma were
identified depression were less likely to report that a vignette character with predicted by social anxiety symptoms as measured by the SAS-A.
symptoms of depression made them feel angry. The present study did not
include a separate measure of “anger” as an emotional response so the
findings are not directly comparable.
5
H. Lynch et al. designed contact with in dividuals with mental disorders. For girls,
more attention should be directed towards their perception of the
This differs somewhat from Calear et al.’s (2011) findings among Australian
stigmatizing attitudes of the general public. They may need
adolescents, which found that personal depression stigma was not
reassurance that they may be over esti mating the level of public
predicted by anxiety but perceived depression stigma was. Our findings
stigma as evidenced by consistent research findings that personal
showed no effect for either depression or social anxiety disorder stigma. It
stigma is lower in the population than perceived stigma.
is important to note that Calear et al.’s (2011) study used a measure of
general anxiety, whereas the present study measured social anxiety, the 4.3. Limitations
measures of stigma were also different. These methodo logical differences
may account for the differences in findings. However it is also important to This study’s findings must be interpreted in light of some limitations.
note that few participants in the present study had high scores on the SAS- Firstly, participants responded to vignette characters, for ethical reasons it
A, as they were drawn from the general popu lation. It is possible that is not possible to elicit responses to real peers, however there are likely to
adolescents with average/low scores on the SAS-A, are not sufficiently be important differences in responses to hypothetical vignette char acters
concerned about negative evaluation and social distancing to have higher versus real peers with mental disorders. A further limitation is that we did
scores on measures of personal and perceived stigma than their peers. It not include a measure of depression symptoms. We did this because our
would be important, therefore, for future research to explore whether these main interest was in social anxiety stigma but this omission has limited our
findings hold for individuals with a clinical diagnosis of social anxiety
insight into the predictors of depression stigma. Another potential limitation
disorder.
relates to the possibility that the adolescents’ re sponses to the stigma
The final aim of the study was to compare stigma levels between
questionnaire were influenced by social desir ability. However, the large
depression and social anxiety. Adolescents reported agreeing with and
standard deviations for the stigma variables
being aware of more stigma towards depression than social anxiety
disorder, although the size of this difference was small. This is
consistent with Anderson et al.’s (2015) finding that young people held
less per sonal stigma towards social anxiety disorder than depression,
as measured by preferred social distance. One interpretation of this
finding is that young people may be more accepting of peers with
social anxiety symptoms and perceive them as more normative.
Another interpretation is that young people know may know less about
social anxiety disorder, as evidenced by the fact that it was identified
by less than half of the participants, and therefore be less likely to
associate it with negative mental health stereotypes.

4.2. Implications

The findings of the study reinforce and confirm the recommenda

tions from other studies of the importance of measuring personal and
perceived stigma separately because of their conceptual distinction. A
further related implication is that the stigma associated with different
mental disorders should be measured separately. Further research is
needed on the stigma associated with social anxiety disorder as our
findings cannot determine whether the lower levels of personal stigma
are due to more positive attitudes or not recognizing the symptoms as
related to a mental disorder. Implications of the gender differences that
were found should be considered when designing mental health
literacy interventions for adolescents. The finding that boys had higher
levels of personal stigma suggests that interventions designed to
reduce stigma may be particularly important for them, these can
include education about the nature of mental disorders or well-
 Journal of Affective Disorders xxx (xxxx) xxx Limitations of the study

indicates varied responses among participants, and suggests Participants responded to vignette characters, for ethical reasons it
adolescents here did not feel compelled to answer a certain way. is not possible to elicit responses to real peers, however there are
likely to be important differences in responses to hypothetical vignette
4.4. Conclusion characters versus real peers with mental disorders. A further potential
limitation relates to the possibility that the adolescents’ responses to
The study set out to increase our understanding of stigma the stigma questionnaire were influenced by social desirability.
associated with social anxiety disorder in adolescence and to provide
information of use to those who wish to design mental health literacy Declaration of Competing Interest
interventions designed to increase knowledge and reduce stigma. The
findings of the study indicate that stigma towards social anxiety The authors declare no conflict of interest
disorder is signifi cantly lower than towards depression and that there
are important gender differences in both personal and perceived Acknowledgements
stigma that need to be considered in intervention design.
The authors wish to express their gratitude to the schools and stu
Author contributions
dents who took part in this research project.
HL: conceptualization, choice of methods, data collection and anal
Supplementary materials
ysis, original draft.
CM: conceptualization, choice of methods, data collection and
Supplementary material associated with this article can be found, in
analysis, review and revision of drafts.
EH: conceptualization, choice of methods, supervision, review and the online version, at doi:10.1016/j.jad.2020.11.073.
revision of drafts. References
Role of the Funding source American Psychiatric Association, 2013. Diagnostic and Statistical Manual of Mental
Disorders, (5th ed.). American Psychiatric Association, Washington, DC.
This research project received no external funding. Anderson, K.N., Jeon, A.B., Blenner, J.A., Wiener, R.L., Hope, D.A., 2015. How
people evaluate others with social anxiety disorder: a comparison to depression
and general mental illness stigma. Am. J. Orthopsychiat. 85 (2), 131–138.
Institutional Board Review https://doi.org/ 10.1037/ort0000046.
Belfer, M.L., 2008. Child and adolescent mental disorders: the magnitude of the
Ethics approval was granted by the University College Dublin problem across the globe. J. Child Psychol. Psyc 49 (3), 226–236.
https://doi.org/10.1111/ j.1469-7610.2007.01855.x.
(UCD) Human Research Ethics Committee - Humanities.
6
H. Lynch et al. Jolliffe, D., Farrington, D.P., 2006. Development and validation of the Basic Empathy
Scale. J. Adolescence 29 (4), 589–611. https://doi.org/10.1016/j.
Busby Grant, J., Bruce, C.P., Batterham, P.J., 2016. Predictors of personal, perceived adolescence.2005.08.010.
Jorm, A.F., Wright, A., 2008. Influences on young people’s stigmatising attitudes towards
and self-stigma towards anxiety and depression. Epidemiol. Psych. Sci. 25, 247–
peers with mental disorders: national survey of young Australians and their parents. Brit.
254. https://doi.org/10.1017/S2045796015000220.
J. Psychiat. 192, 144–149. https://doi.org/10.1192/bjp.bp.107.039404.
Calear, A.L., Griffiths, K.M., Christensen, H., 2011. Personal and perceived
Keller, M.B., 2003. The lifelong course of social anxiety disorder: a clinical perspective.
depression stigma in Australian adolescents: magnitude and predictors. J. Affect.
Acta Psychiat. Scand. 108 (417), 85–94. https://doi.org/10.1034/j.1600-0447.108.
Disorders. 129, 104–108. https://doi.org/10.1016/j.jad.2010.08.019.
Calear, A.L., Batterham, P.J., Griffiths, K.M., Christensen, H., 2017. Generalized anxiety s417.6.x.
disorder stigma in adolescents: personal and perceived stigma levels and predictors. Kogan, C.S., Stein, D.J., Maj, M., First, M.B., Emmelkamp, P.M.G., Reed, G.M., 2016.
Stigma and Health 2 (3), 208–215. https://doi.org/10.1037/sah0000046. The classification of anxiety and fear-related disorders in the ICD- 11. Depress
Coles, M.E., Ravid, A., Gibb, B., George-Denn, D., Bronstein, L.R., McLeod, S., Anxiety 33 (12), 1141–1154. https://doi.org/10.1002/da.22530.
2016. Adolescent mental health literacy: young people’s knowledge of
depression and social anxiety disorder. J. Adolescent Health. 58 (1), 57–62.
https://doi.org/ 10.1016/j.jadohealth.2015.09.017.
Corrigan, P.W., Shapiro, J.R., 2010. Measuring the impact of programs that challenge
the public stigma of mental illness. Clin. Psychol. Rev. 30, 907–922.
https://doi.org/ 10.1016/j.cpr.2010.06.004.
Curcio, C., Corboy, D., 2019. Stigma and Anxiety Disorders: a Systematic Review.
Stigma and Health. https://doi.org/10.1037/sah0000183.
Dardas, L.A., Silva, S.G., Smoski, M.J., Noonan, D., Simmons, L.A., 2017. Personal
and perceived depression stigma among Arab adolescents: associations with
depression severity and personal characteristics. Arch. Psychiat. Nurs. 31, 499–
506. https://doi. org/10.1016/j.apnu.2017.06.005.
Deater-Deckard, K., 2001. Annotation: Recent research examining the role of peer
relationships in the development of psychopathology. J. Child Psychol. Psyc.
42, 565–579. https://doi.org/10.1017/S0021963001007272.
Dolphin, L., Hennessy, E., 2014. Adolescents‫ ׳‬perceptions of peers with depression:
an attributional analysis. Psychiatry Research 218, 295–302.
https://doi.org/10.1016/ j.psychres.2014.04.051.
Dolphin, L., Hennessy, E., 2016. Depression stigma among adolescents in Ireland.
Stigma and Health 1, 185. https://doi.org/10.1037/sah0000025.
Festa, C.C., Ginsburg, G.S., 2011. Parental and peer predictors of social anxiety in youth.
Child Psychiat. Hum. D. 42, 291–306. https://doi.org/10.1007/s10578-011-0215-8.
Flannery, K.M., Smith, R.L., 2017. The effects of age, gender, and gender role ideology on
adolescents’ social perspective-taking ability and tendency in friendships. J. Soc. Pers.
Relat. 34 (5), 617–635. https://doi.org/10.1177/0265407516650942.
Hanlon, H.R., Swords, L., 2019. Overthinkers, attention-seekers and wallflowers: peer
perceptions of clinical anxiety disorders in adolescence. J. Pub. Ment. Health 18
(1), 4–13. https://doi.org/10.1108/JPMH-07-2018-0049.
Heary, C., Hennessy, E., Swords, L., Corrigan, P., 2017. Stigma towards mental
health problems during childhood and adolescence: theory, research and
intervention approaches. J Child Fam Stud 26 (11), 2949–2959.
https://doi.org/10.1007/ s10826-017-0829-y.
Hertzman, C., Boyce, T., 2010. How experience gets under the skin to create
gradients in developmental health. Ann. Rev. Publ. Health. 31, 329–347.
https://doi.org/ 10.1146/annurev.publhealth.012809.103538.
 Journal of Affective Disorders xxx (xxxx) xxx severity, age of onset, continuity and co-morbidity of DSM disorders. Psychol.
Med. 45 (2), 345–360. https://doi.org/10.1017/S0033291714001469.
Kutcher, S., Wei, Y., Coniglio, C., 2016. Mental health literacy: past, present and future.
Pallant, J., 2001. SPSS survival manual: a step by step guide to data analysis using
SPSS for Windows (versions 10 and 11): SPSS student version 11.0 for Windows.
Can. J. Psychiat. 61 (3), 154–158. https://doi.org/10.1177/0706743715616609.
Open University Press, Milton Keynes.
Masten, C.L., Eisenberger, N.I., Borofsky, L.A., Pfeifer, J.H., McNealy, K., Mazziotta,
Schafer, J.L., 1999. Multiple imputation: a primer. Stat. Methods. Med. Res. 8, 3–15.
J.C., Dapretto, M., 2009. Neural correlates of social exclusion during
https://doi.org/10.1177/096228029900800102.
adolescence: understanding the distress of peer rejection. Soc. Cogn. Affect. Storch, E.A., Masia-Warner, C., Dent, H.C., Roberti, J.W., Fisher, P.H., 2004. Psychometric
Neur. 4, 143–157. https://doi.org/10.1093/scan/nsp007. evaluation of the Social Anxiety Scale for Adolescents and the Social Phobia and Anxiety
Merikangas, K.R., He, J.P., Burstein, M., Swanson, S.A., Avenevoli, S., Cui, L., ..., Inventory for Children: construct validity and normative data. J. Anxiety Disord. 18, 665–
Swendsen, J., 2010. Lifetime prevalence of mental disorders in US adolescents: 679. https://doi.org/10.1016/j.janxdis.2003.09.002.
results from the National Comorbidity Survey Replication- Adolescent
Tabachnick, B.G., Fidell, L.S., 2001. Computer-assisted research design and analysis.
Supplement (NCS-A). J. Am. Acad. Child. Psy. 49 (10), 980–989.
Allyn and Bacon, Boston.
https://doi.org/10.1016/j. jaac.2010.05.017.
Wacker, R., Bolte,¨ S., Dziobek, I., 2017. Women know better what other women think
McKeague, L., Hennessy, E., O’Driscoll, C., Heary, C., 2015a. Retrospective
and feel: gender effects on mindreading across the adult life span. Front. Psychol.
accounts of self-stigma experienced by young people with attention-
8, 1324. https://doi.org/10.3389/fpsyg.2017.01324.
deficit/hyperactivity disorder (ADHD) or depression. Psychiatr. Rehabil. J. 38 (2),
World Health Organization, 2017. Depression and other common mental disorders:
158. https://doi.org/ 10.1037/prj0000121.
global health estimates. World Health Organization (No. WHO/MSD/MER/2017.2).
McKeague, L., Hennessy, E., O’Driscoll, C., Heary, C., 2015b. Peer Mental Health
https://apps.who.int/iris/bitstream/handle/10665/254610/WHO-MSD-MER-2017. 2-
Stigmatization Scale: psychometric properties of a questionnaire for children
eng.pdf.
and adolescents. Child Adol. Ment. H. 20, 163–170. https://doi.org/10.1111/ World Health Organization, 2018. International Statistical Classification of Diseases and
camh.12088.
Related Health Problems. World Health Organization, Geneva (11th Revision).
Moses, T., 2014. Determinants of mental illness stigma for adolescents discharged
Wright, A., Jorm, A.F., 2009. Labels used by young people to describe mental
from psychiatric hospitalization. Soc. Sci. Med 109, 26–34.
disorders: factors associated with their development. Aust NZ J Psychiat 43, 946–
https://doi.org/10.1016/j. socscimed.2013.12.032.
955. https:// doi.org/10.1080/00048670903179129.
Nearchou, F.A., Bird, N., Costello, A., Duggan, S., Gilroy, J., Long, R., . . ., Hennessy,
Yap, M.B.H., Reavley, N., Mackinnon, A.J., Jorm, A.F., 2013. Psychiatric labels and
E., 2018. Personal and perceived public mental-health stigma as predictors of
help-seeking intentions in adolescents. J. Adolescence 66, 83–90. https://doi.org/ other influences on young people’s stigmatizing attitudes: findings from an
10.1016/j.adolescence.2018.05.003. Australian national survey. J. Affect. Disorders 148 (2-3), 299–309.
Ormel, J., Dennis, Raven, van Oort, Floor, Hartman, C.A., Reijneveld, S.A., Ren´e https://doi.org/10.1016/j. jad.2012.12.015.
Veenstra, Vollebergh, W.A.M., Jan Buitelaar, Verhulst, F.C., Oldehinkel, A.J.,
2015. Mental health in Dutch adolescents: a TRAILS report on prevalence,