The Longitudinal Association Between Body Image Dissatisfaction, Social Anxiety, and Fear of Negative Evaluation in Adolescents
Table Of Contents
Project Abstract
<p> <b>Abstract
</b></p><p>Adolescents with body image dissatisfaction experience more anxiety than their peers who are
more satisfied with their body. This is problematic given that adolescents who experience these
concerns have a greater likelihood of later developing other mental health disorders and have
more disordered eating cognitions and behaviour. For this reason, I investigated how body image
dissatisfaction, social anxiety, and fear of negative evaluation were related to one another.
Participants included 527 adolescents (301 girls; aged 15 to 19 years; 83.1% White) who were
accessed annually over 4 years (Grade 10 to one-year post high school) using the
Multidimensional Anxiety Scale for Children, the Brief Fear of Negative Evaluation scale-II, and
validated questions to assess body image dissatisfaction. A developmental cascade model was
used to examine direct and indirect effects between the study variables. Results indicated two
significant indirect paths; body image dissatisfaction to social anxiety via fear of negative
evaluation and body image dissatisfaction to fear of negative evaluation via social anxiety. Direct
effects included a reciprocal positive association between body image dissatisfaction and social
anxiety in mid-adolescence and a reciprocal positive association between social anxiety and fear
of negative evaluation across adolescence. Lastly, there was a positive association from body
image dissatisfaction to fear of negative evaluation across adolescence. These results suggest that
adolescents with low body image dissatisfaction are likely to experience greater fear and anxiety
regarding social interaction. This study emphasizes the need to target adolescents with body
image intervention programs to reduce the experience of psychopathology
<br></p>
Project Overview
<p>1.0 INTRODUCTION</p><p>1.1 BACKGROUND STUDY</p><p>
Anxiety disorders include “excessive fear and anxiety and related behavioral
disturbances” (American Psychiatric Association, 2013, p. 189). Fear encompasses the emotional
response to a threat that is either present or absent, whereas anxiety is the concern over a threat
occurring in the future (American Psychiatric Association, 2013). With anxiety disorders,
individuals develop a preconscious attentional bias toward a stimulus which is deemed
threatening (Craske et al., 2009). Consequently, individuals have an elevated sensitivity to the
feared object(s) (Craske et al., 2009). Anxiety can be present in children. For example, a child
may exhibit separation anxiety when their parent leaves the room (Rynn, Vidair, & Blackford,
2012). Of all the mental disorders, anxiety disorders are the most frequent class of psychiatric
disorders that children and adolescents are diagnosed with (Rynn et al., 2012).
Anxiety is a global concern with one in 14 adults experiencing an anxiety disorder at any
given time (Baxter, Scott, Vos, & Whiteford, 2013). The prevalence of anxiety disorders in
youth is also high, with an estimation that 204,400 Canadian youth between the ages of 4- and
17-years affected (prevalence of 3.8%; Waddell, Shepherd, Schwartz, & Barican, 2014).
Childhood and adolescence is a critical developmental period where youth are at risk of
developing symptoms of anxiety which could range anywhere from mild symptoms to
diagnosable anxiety disorders (Beesdo, Knappe, & Pine, 2009). These findings are disconcerting
given that youth who develop anxiety disorders are more likely to be affected by subsequent
anxiety disorders (Beesdo et al., 2009; Clark, Smith, Neighbors, Skerlec, & Randall, 1994,
Kashani & Orvaschel, 1990) and other mental illnesses throughout their life (Beesdo et al.,
2009). Specifically, in children aged 8- through 17-years, generalized anxiety disorder was found
to be the most common psychiatric disorder (Kashani & Orvaschel, 1990). Anxiety regarding
social fears, interpersonal concerns, and personal adequacy were highest in late adolescence
(Kashani & Orvaschel, 1990). As well, it is more likely for older adolescents to experience social
anxiety compared to younger adolescents (Burstein et al., 2011), making the sample of the
present study—mid to late adolescence—particularly relevant.
There are a variety of negative outcomes that arise from youth experiencing a mental
health problem. Anxiety negatively impacts interpersonal and intrapersonal behaviour in
adolescents (Kashani & Orvaschel, 1990). Kashani and Orvaschel (1990) found that anxious 17-
year-olds (compared to non-anxious 17-year-olds) had worse social relationships and more
behavioural concerns, mood problems, somatic complaints, and school difficulties. Having more
anxiety symptoms is related to poorer health-related quality of life (Raknes et al., 2017). Given
the decreased quality of life that accompanies anxiety symptoms, there is a need to improve
mental health interventions and implement prevention initiatives that target anxious adolescents.
Anxiety disorders are differentiated from one another based on the situation or object that
contributes to the anxious or fearful response. A few examples of anxiety disorders include
separation anxiety (fear or anxiety experienced when separated from one’s attachment figure),
specific phobia (fear or anxiety toward a particular object or situation), and of particular
relevance to the present study, social anxiety (American Psychiatric Association, 2013).
With social anxiety disorder, the fear or anxiety experienced by an individual is related to
social situations or interpersonal interaction. The cognitive component of this disorder often
involves worry over being embarrassed, humiliated, or rejected by others (American Psychiatric
Association, 2013). The presentation of social anxiety, thus, has a strong social evaluative
component
Social anxiety exists on a spectrum, such that individuals can experience symptoms of
social anxiety without meeting the criteria for a formal diagnosis for social anxiety disorder. The
12-month prevalence rates for children and adolescents for social anxiety disorder are
comparable to adults which is around 7% (American Psychiatric Association, 2013). Stein,
Walker, and Forde (1994) found that 33.3% of Canadian adults were “much more nervous than
other people” in at least one social situation, although these participants did not necessarily have
social anxiety disorder (p. 410). This suggests that it is relatively common to experience anxiety
in social situations; however, it is much less common to have social anxiety disorder.
It is well known that the diagnosis of social anxiety results in a greater risk of other
mental health disorders and overall life impairments; however, even individuals with subclinical
expressions of social anxiety have been found to experience these impairments (Fehm, Beesdo,
Jacobi, & Fiedler, 2008). The participants in the present study were assessed in terms of social
anxiety symptoms and were not required to meet the criteria for social anxiety disorder to be
considered.
Regarding body image, body image dissatisfaction has been found to be positively related
to anxiety (e.g., Cruz-Sáez, Pascual, Salaberria, & Echeburúa, 2015; Dooley, Fitzgerald, &
Giollabhui, 2015; Duchesne et al., 2017). In one study, anxious 17-year-olds had a worse selfconcept than same-aged adolescents without anxiety (Kashani & Orvaschel, 1990). Thus, it
appears that those with anxiety are more likely to experience body image dissatisfaction. Body
image is defined as “one’s thoughts, perceptions, and attitudes about their physical appearance”
which includes what one believes about their appearance and how one thinks and feels about
their body (Body Image & Eating Disorders, 2018). Body dissatisfaction usually has negative
impacts on a person’s life. For example, those with body image dissatisfaction have been found
to experience distress and/or attempt to alter their appearance (Cash, 1996). Neumark-Sztainer,
Paxton, Hannan, Haines, and Story (2006) found that adolescent boys and girls were more likely
to diet and engage in unhealthy weight control behaviour if they were dissatisfied with their
body. Men and women in late adulthood with body image dissatisfaction were more likely to
experience anxiety and depression, and middle-aged men were more likely to have problematic
social and sexual functioning if they were dissatisfied with their body (Davison & McCabe,
2005). Conversely, individuals who were satisfied with their body image were less likely to
experience anxiety and to have problematic internet use, and were more likely to have good selfesteem (Cash, 1996).
Specific to youth, body image dissatisfaction has been found to be related to a variety of
psychological well-being factors including more distress, depression, and anxiety, and worse
self-esteem and self-concept. In their review of the literature on body image in boys, Cohane and
Pope (2001) found that body dissatisfaction in boys (under 18 years) was often associated with
distress (e.g., impaired self-concept and self-esteem). Similarly, Kostanski and Gullone (1998)
found that body image dissatisfaction was negatively correlated with self-esteem in adolescents
between the ages of 12 and 18. Body image dissatisfaction was also related to depression and
anxiety, such that the more body image dissatisfaction that an adolescent reported, the more
anxiety and depression they were likely to have (Duchesne et al., 2017; Kostanski & Gullone,
1998). Thus, the more concerned adolescents are about their body image, the less likely they are
to have good self-esteem and the more likely they are to experience psychological distress (i.e.,
depression and anxiety). Given that body image dissatisfaction is related to various
psychological disturbances in youth, adolescence is a prime time to investigate this variable.
With an understanding that body image dissatisfaction tends to be relatively consistent
throughout adulthood (Davison & McCabe, 2005), it is important that adolescent body image is
acknowledged rather than waiting for these concerns to negatively impact individuals throughout
their life.
Specifically concerning anxiety, more body image dissatisfaction has been shown to be
associated with more anxiety (e.g., Di Blasi et al., 2015; Duchesne et al., 2017). Halliwell and
Dittmar (2003) found that women (mean age of 31 years) who were exposed to thin models in
the media had more body-focused anxiety compared to those who were exposed to average-sized
models or no models. This suggests that body image in the media impacts anxiety. Given that we
are regularly surrounded by advertising in various forms, being exposed to thin models is often
unavoidable making it likely for anxiety to be experienced. When individuals experience anxiety
about their health they were more likely to engage in body checking (Hadjistavropoulos &
Lawrence, 2007), which provides further support to the argument that anxiety and body image
are linked. It is possible that those who experience body image dissatisfaction are more vigilant
of their body, leading them to experience anxiety in social situations since they are not feeling
confident with their body.
Gaining a better understanding of the relation between body image dissatisfaction and
anxiety in adolescents is important for informing, and consequently improving, the mental health
of adolescents. A better understanding can come from investigating the potential impact that
certain variables have on body image dissatisfaction and social anxiety. For instance, children
who were rejected by their peers have been found to experience increased negative affect and an
increased likelihood to engage in maladaptive social behaviour (Nesdale & Lambert, 2008). It
makes sense that children who were rejected by their peers are more likely to experience
negative outcomes. However, even children who were not actually rejected, but anticipate
rejection (i.e., children high on rejection sensitivity; Park 2007), were more likely to experience
negative outcomes. Downey, Lebolt, Rincón, and Freitas (1998) found that children (in Grades 5
to 7) who angrily expected rejection became more distressed compared to children who were less
afraid of being rejected. Downey et al. also found that children who were high on rejection
sensitivity had more social conflicts (with peers and teachers), were more disruptive and
oppositional, and were less engaged in school. Being sensitive to rejection appears to be related
to a variety of negative outcomes. It is possible that those who are most afraid of being rejected
are more likely to experience social anxiety given that social anxiety involves being fearful of
being rejected by others.
In the present study, I examined the relation between body image dissatisfaction, social
anxiety, and fear of negative evaluation. I investigated whether fear of being negatively
evaluated influenced the relation between body image dissatisfaction and social anxiety in
adolescents, with the aim of informing current body image intervention strategies. In the
literature review, an overview of the literature on the variables being studied is discussed which
includes the impact and prevalence of body image dissatisfaction and social anxiety, as well as
the relation between these two variables. Fear of negative evaluation is explained and considered
in relation to body image dissatisfaction and social anxiety. Further, any longitudinal studies
involving these variables are discussed. A detailed description of the study’s objectives and
methodology follows. The results are then presented and discussed as well as the future
implications and the contribution of my research.
<br></p>