See discussions, stats, and author profiles for this publication at: https://www.researchgate.net/publication/11899417
Social anxiety in college students
Article in Journal of Anxiety Disorders · May 2001
DOI: 10.1016/S0887-6185(01)00059-7 · Source: PubMed
CITATIONS
READS
79
32,130
4 authors:
Christine L Purdon
Martin M Antony
University of Waterloo
Toronto Metropolitan University
108 PUBLICATIONS 6,006 CITATIONS
350 PUBLICATIONS 18,475 CITATIONS
SEE PROFILE
SEE PROFILE
Sandra Monteiro
Richard P Swinson
McMaster University
McMaster University
88 PUBLICATIONS 1,798 CITATIONS
240 PUBLICATIONS 15,147 CITATIONS
SEE PROFILE
SEE PROFILE
Some of the authors of this publication are also working on these related projects:
Technology and Cognition View project
Learning Taken Out of Context View project
All content following this page was uploaded by Christine L Purdon on 20 October 2018.
The user has requested enhancement of the downloaded file.
Anxiety Disorders
15 (2001) 203 ± 215
Social anxiety in college students
Christine Purdona,b,*, Martin Antonya,c, Sandra Monteiroa,
Richard P. Swinsona,c
a
Anxiety Treatment and Research Centre, St. Joseph's Hospital, Hamilton, Ontario, Canada
b
Department of Psychology, University of Waterloo, Waterloo, Ontario, Canada N2L 3G1
c
Department of Psychiatry and Behavioural Neuroscience, McMaster University,
Hamilton, Ontario, Canada
Received 7 October 1998; accepted 14 June 1999
Abstract
Individuals with social phobia often hold erroneous beliefs about the extent to which
others experience symptoms of social anxiety and the ways in which others evaluate people
who appear to be anxious. The purpose of this study was to: (a) provide normative data on
the frequency with which individuals in a nonclinical sample experience particular
symptoms of social anxiety (e.g., sweating, shaking, etc.); (b) to examine how the
perception of anxiety in others influences participants' immediate impressions of various
personal characteristics (e.g., intelligence, attractiveness, etc); and, (c) investigate the
relationship between social anxiety and perceptions regarding others who appear to be
anxious. Eighty-one undergraduate students completed self-report measures of social
anxiety and social desirability, and then rated the degree to which their impressions of
various personal characteristics were influenced when another individual was perceived to
be anxious. Results suggested that the vast majority of individuals experience symptoms of
anxiety in social situations from time to time. In addition, individuals who themselves
reported elevated social anxiety were more likely than individuals less socially anxious to
judge others who appear anxious to have less strength of character and to be less attractive
and more compassionate compared to others who do not appear anxious. Clinical
implications of these results are discussed. D 2001 Elsevier Science Inc. All rights reserved.
Keywords: Social anxiety; Anxiety symptoms; Social phobia
* Corresponding author. Department of Psychology, University of Waterloo, Waterloo, Ontario,
Canada N2L 3G1. Tel.: +1-519-888-4567, ext. 3912; fax: +1-519-746-8631.
E-mail address: clpurdon@watarts.uwaterloo.ca (C. Purdon).
0887-6185/01/$ ± see front matter D 2001 Elsevier Science Inc. All rights reserved.
PII: S 0 8 8 7 - 6 1 8 5 ( 0 1 ) 0 0 0 5 9 - 7
204
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
1. Introduction
Social phobia is an anxiety disorder that is characterized by the strong desire
to make a favorable impression of oneself on others, in conjunction with a
marked insecurity about one's ability to do so. Individuals with social phobia
avoid situations in which there is potential for negative evaluation by others, or
endure such situations with great anxiety and distress (American Psychiatric
Association, 1994; Clark & Wells, 1995; Rapee & Heimberg, 1997). According
to cognitive ±behavioral models of social phobia, social anxiety is maintained by
excessively high standards for social performance (e.g., ``I must not let anyone
see I am anxious''), a tendency to assume that others view oneself as inadequate
(e.g., boring, peculiar, unattractive, etc.), and a tendency to assume that others'
beliefs about oneself are true (Clark & Wells, 1995; Rapee & Heimberg, 1997).
As a result of these beliefs and assumptions, individuals with social phobia tend
to: (1) report a high frequency of negative self-statements, (2) negatively evaluate
the quality of their social performance, (3) notice what went wrong in a social
interaction rather than what went right, (4) be preoccupied with how others are
evaluating them, and (5) engage in excessive self-monitoring of their presentation
to others, including attention to physiological symptoms of their anxiety (Clark &
Wells, 1995; Mattick, Page, & Lampe, 1995; Rapee & Heimberg, 1997;
Scholing, Emmelkamp, & van Oppen, 1996; Wells, 1997). According to Clark
and Wells (1995) individuals with social phobia also tend to rely on internal
``feeling states'' as a means of judging whether or not a social interaction is going
well. That is, such individuals tend to assume that if they feel anxious in a social
situation, it is because they are not performing well. Nonphobic individuals, on
the other hand, will often test their interpretation of a social situation by various
strategies, such as seeking out further eye contact to determine if another is
genuinely uninterested in what they are saying, for example, and are thus able to
appraise their performance more realistically.
Socially anxious individuals, then, tend to construct highly negative images of
their performance in social situations which contribute substantially to anticipatory anxiety as well as negative postevent processing (Clark & Wells, 1995;
Rapee & Heimberg, 1997; Wells, 1997). This anxiety results in the use of safetyseeking behaviors in social situations (e.g., mentally practicing what one is going
to say next in a conversation, wearing high collared sweaters to mask signs of
blushing, gripping a glass tightly with one hand to avoid spilling) that can
actually cause or exaggerate the feared symptoms (e.g., planning one's next
sentence makes it hard to keep up with the conversation itself, wearing sweaters
makes one hot thereby increasing flushing, a tight grip can actually increase
tremor). In combination with information processing biases in evaluating social
performance, such behaviors prevent learning of new information about the
consequences of the feared event (e.g., spilling something does not result in
widespread social rejection). At the same time, concern about the importance of
not showing anxiety results in excessive monitoring of bodily sensations, and
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
205
slight fluctuations are detected and result in an increase in anxiety. Cognitive
behavioral treatment of social phobia targets erroneous beliefs and assumptions
about social situations, as well as avoidance and safety-seeking behaviors.
The belief that symptoms of anxiety in oneself will be perceived by others as a
sign of weakness, incompetence, mental instability, stupidity, etc. and lead to
humiliation or embarrassment is a cardinal feature of social anxiety and accounts
for much of the negative self-evaluation in social situations reported by social
phobic individuals (APA, 1994; Scholing & Emmelkamp, 1993; see also Clark &
Wells, 1995, for a review of clinical anecdotal evidence). Individuals with social
phobia also have a tendency to overestimate the extent to which symptoms of
anxiety are visible to others (Alden & Wallace, 1995; Bruch, Borsky, Collins, &
Berger, 1989; McEwan & Devins, 1983). Socially anxious individuals, then, are
in the position of believing that their anxiety is a severe social handicap, whilst
having a distorted image of the extent to which their symptoms of anxiety are
observable. It would seem, then, that if individuals were not so concerned about
the consequences of exhibiting symptoms of anxiety in social situations, their
anxiety might decrease substantially. Correction of erroneous beliefs about the
extent to which symptoms of anxiety are experienced within the general
population, as well as the extent to which symptoms of anxiety are perceived
negatively by others might, then, be helpful in treating social anxiety.
However, to what extent are negative beliefs about the impact of showing
signs of anxiety accurate, and to what extent is social anxiety an aberrant
response? To date, these questions have not been addressed in the literature.
Investigations of social anxiety in nonclinical samples have focused on general
behavior, such as social avoidance and general feelings of anxiety. Such studies
reveal that 40% of individuals consider themselves to be ``shy,'' and that 90% of
individuals report having had periods in their life when they were shy (Zimbardo,
1977). Shyness, in general, is defined by social inhibition and anxiety, and is
considered by some to fall on a continuum with social phobia (Rapee &
Heimberg, 1997; Turner, Beidel, & Townsley, 1990). Similarly, other research
has observed subclinical social anxiety to be quite prevalent in the general
population, with 50% to 61% of individuals reporting social anxiety in at least
one situation (Hofmann & Roth, 1996; Stein, Walker, & Forde, 1994). However,
information about the frequency with which individuals in nonclinical samples
experience specific symptoms of social anxiety, such as stammering, blushing,
shaking, etc., is not currently available. As well, no studies to date have examined
the degree to which one's immediate impression of another person is influenced
by detecting signs of anxiety. It would seem, then, that the majority of individuals
do experience social anxiety from time to time and therefore may have a neutral,
if not sympathetic, response to noticing anxiety in others. If this is the case, it
might be helpful in treatment to provide this data in order to help alleviate
concerns about negative evaluation.
The purpose of this study was to: (a) obtain normative data on the frequency
of symptoms of social anxiety in a nonclinical sample, (b) assess the degree to
206
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
which anxiety influences people's immediate impression of others, and (c)
investigate the relationship between social anxiety and immediate impressions
of others who appear to be anxious. Participants were administered self-report
measures of social anxiety, a measure of the frequency of particular manifestations of social anxiety, a questionnaire examining the degree to which social
anxiety influenced immediate impressions of various characterological features,
and a measure of social desirability. Given the prevalence of situational social
anxiety and subclinical ``shyness'' in the general population, it was hypothesized
that symptoms of anxiety would be quite common in a nonclinical sample. It was
also hypothesized that, given the expected prevalence of symptoms of social
anxiety in nonclinical samples, the immediate impressions of others formed by
individuals low in social anxiety would not be negatively influenced by the
observation that that individual was anxious. Finally, given that the primary
preoccupation of individuals with social phobia is a concern with failing to meet
the high standards of others (Rapee & Heimberg, 1997), we expected that
individuals high in social anxiety would perceive anxiety symptoms not as a
problem in and of themselves, but rather as a problem only to the extent that they
resulted in a failure to meet the high expectations of performance imposed by
others. Therefore, it was hypothesized that the immediate impressions of
individuals high in social anxiety would also not be negatively influenced by
the observation that an individual is anxious.
2. Method
2.1. Participants
Participants were 81 undergraduate student volunteers with a mean age of 25
(S.D. = 6.58). The sample consisted of 60 women and 21 men. Participants were
solicited from first and second year undergraduate social sciences classes, and
were administered questionnaire packets which they completed at the end of class.
2.2. Measures
2.2.1. Social Phobia Scale (SPS) and Social Interaction Anxiety Scale (SIAS)
(Mattick & Clarke, 1998)
The SPS and SIAS are both 20-item self-report measures of social anxiety.
The SPS assesses anxiety experienced when anticipating being observed or
actually being observed by other people and when undertaking certain activities
in the presence of others (e.g., public speaking, eating, or writing). The SIAS
assesses anxiety associated with social interaction situations (e.g., initiating
conversations with others). Both instruments have strong internal consistency
and appear to be reliable in discriminating individuals with social anxiety from
those with other anxiety disorders (Brown, Turovsky, Heimberg, Juster, Brown,
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
207
& Barlow, 1997; Mattick & Clarke, 1998). These scales were included in order to
examine the relationship between social anxiety and judgments of others with
social anxiety.
2.2.2. Symptoms of social anxiety scale (SASS)
The SASS was developed for use in this study. It lists 24 symptoms of anxiety,
including symptoms that are experienced internally (e.g., heart palpitations,
dizziness) and symptoms that might be noticed by others (e.g., blushing, sweating,
shaky voice). The list of items was designed to encompass the range of physiological correlates of anxiety identified in the existing literature, and was shown to
experts in anxiety disorders who rated it informally for comprehensiveness. The
final list of items comprising the SASS was based on these ratings. Respondents
rate the frequency with which they experience each symptom in social situations
using a Likert scale ranging from 0 (Never) to 4 (Always). This measure was
included in order to obtain normative information on the frequency of social
anxiety symptoms in a nonclinical sample. The relationship between scores on this
measure and judgments about others with social anxiety were also examined.
2.2.3. Negative Evaluation Questionnaire (NEQ)
The NEQ was also developed for use in this study. The questionnaire lists
eight different qualities (intelligence, attractiveness, strength of character, compassion, leadership abilities, ambition, reliability, and mental health). Respondents are asked to rate how the presence of social anxiety in another person
would influence the participant's perception of each of these qualities in that
individual. The general instructions were: ``Research indicates that the degree to
which an individual appears calm vs. anxious can very much influence how she/
he is perceived by others. We are interested in how your immediate impression of
another individual is influenced when you notice that he/she is anxious. When
you notice that someone you are with is anxious, to what extent does this affect
your immediate impression of their . . .[compassion, strength of character, etc.].''
Respondents rate each quality using a five-point Likert scale ranging from 1 (I
would think they are less. . .[compassionate, intelligent, etc.]) to 3 (It would not
change my opinion of their. . .) to 5 (I would think they are more. . .). These
qualities were selected on the basis that they represent separate and distinct
characteristics that individuals use to judge their liking of another (Ernulf &
Innala, 1993; Rubin, 1970). The items were intended to be examined separately.
This questionnaire was included to test the main hypotheses of the study.
2.2.4. Marlowe± Crowne Social Desirability Scale (MCSDS)
The MCSDS is a 33-item self-report measure of items reflecting culturally
acceptable and approved behaviors which are, at the same time, relatively unlikely
to occur. High scores on this measure are interpreted as representing a prosocial
response bias. This measure has strong demonstrated reliability and validity
(Crowne & Marlowe, 1960) and has been widely used to assess and control for
208
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
prosocial response biases (Millham & Jacobsen, 1978). This measure was included
in order to control for potential prosocial response biases that might be present in
evaluations of the influence of anxiety on individuals' impressions of others.
3. Results
Means and standard deviations for the SIAS, SPS, and MCSDS are presented
on Table 1. These values are comparable to published means and standard
deviations from nonclinical samples. In order to determine the degree to which
participants reported symptoms of anxiety in social situations, we examined the
mean frequency ratings for each SASS item. Means and standard deviations for
individuals scoring in the top 25th percentile on both the SIAS (score of 30 or
more) and the SPS (score of 17 or more) (i.e., individuals high in social anxiety)
were also obtained. Finally, the percentage of the total sample reporting each
symptom at least ``rarely'' were also obtained. These are reported in Table 2.
The ratings of the high anxiety group were generally higher than the total
group mean on all items. It would have been desirable to conduct a multivariate
analysis of variance (MANOVA) examining differences between those in the
high anxiety group to the rest across all items at once to determine which best
distinguished the groups. However, given the small size of the high anxiety group
(n = 15), this was not feasible. Instead a t test comparing the high anxiety group to
the rest on total SASS scores was conducted. The high anxiety group had
significantly higher scores on the SASS than did the others (t(1,79) = 7.32,
P < .001, suggesting that physiological symptoms of anxiety coexist with fear of
negative evaluation in performance and social situations in nonclinical samples.
All symptoms on the SASS were experienced at least ``rarely'' by at least 13% of
the participants. A total of 11 of the 24 symptoms were experienced ``rarely'' to
``sometimes.'' ``Butterflies'' in the stomach, general tension, desire to avoid a
situation, trouble expressing oneself, and blushing were the five most frequently
experienced. The five least frequently experienced were tingling fingertips,
blurred or distorted vision, numbness in limbs, dizziness, and nausea.
In order to investigate the influence of perceiving social anxiety on individuals' self-reported impressions of those exhibiting it, scores on the NEQ were
Table 1
Means and standard deviations of the SIAS, SPS, and MCSDS
Scale
Mean
S.D.
SIAS
SPS
MCSDS
22.38
14.38
14.19
15.40
12.91
5.08
N's vary from 73 to 81 due to missing data; SIAS = Social Interaction Anxiety Scale, SPS = Social
Phobia Scale, MCSDS = Marlowe ± Crowne Social Desirability Scale.
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
209
Table 2
Means and standard deviations of frequency ratings for each SASS symptom of anxiety
All participants
Socially anxious
participants
% Reporting Sx
at least once
Symptom
Mean
S.D.
Mean
S.D.
%
1. Hot or cold flashes
2. Heart palpitations
3. Chest pain, pressure
4. Blushing
5. Stammering
6. Numbness in limbs
7. Tingling fingertips
8. Trouble expressing self
9. Shortness of breath
10. General tension
11. Wanting to escape
12. Dizziness
13. Wanting to avoid a situation
14. Blurred or distorted vision
15. Nausea
16. ``Butterflies'' in stomach
17. ``Lump'' in throat
18. Sweating
19. Wobbly or rubber legs
20. Dry throat
21. Shaking hands or knees
22. Smiling, laughing,
or talking uncontrollably
or inappropriately
23. Shaky voice
24. Feeling disoriented
or confused
0.84
1.14
0.51
1.35
0.90
0.37
0.32
1.48
0.58
1.49
1.07
0.44
1.43
0.33
0.47
1.56
0.95
1.25
0.65
0.89
1.12
1.09
0.95
0.97
0.76
0.96
0.96
0.71
0.77
1.07
0.93
1.12
1.17
1.00
1.07
0.71
0.82
0.96
1.00
1.01
0.88
0.97
1.17
1.03
1.67
2.07
0.93
1.67
1.73
0.87
0.73
3.00
1.67
2.73
2.53
1.27
2.60
1.20
1.97
2.20
1.67
1.47
1.13
1.73
2.47
2.20
1.18
1.16
0.88
1.29
1.22
1.06
1.16
0.76
1.18
1.16
1.30
1.44
1.18
1.08
0.96
1.01
1.05
1.06
0.99
1.16
0.92
0.94
51.90
71.60
35.80
75.30
56.80
24.70
21.00
81.50
35.80
79.00
56.80
21.00
77.80
21.20
28.40
85.20
54.30
72.80
42.00
55.60
58.00
65.40
1.01
0.64
1.09
1.00
2.40
1.93
1.06
1.28
59.30
35.80
N = 81. High Anxious n = 15. SASS = Social Anxiety Symptoms Scale.
examined. Although five-point Likert scales were used in this measure, examination of item distributions revealed a much more dichotomous picture. That is,
for all but two items, the vast majority of participants (95%+) tended to report
either no change in their immediate impressions of that quality, or a more
negative evaluation of that quality. In two items, compassion and ambition, a
minority of individuals (21% and 16%, respectively) reported that they thought
someone who exhibits social anxiety would be more compassionate or ambitious.
The item variance was, then, very low. In Table 3, items are presented with
frequencies of individuals reporting that their immediate impression of that
quality would be negatively influenced (i.e., a score of 1 or 2), that their
impression of that quality would not change (i.e., a score of 3), or, where
applicable, that their impression of that quality would be positively influenced
(i.e., a score of 4 or 5). More than half of the participants reported that their
210
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
Table 3
Frequency and percentage of individuals reporting negative, positive, or no change in their immediate
impressions of each characteristic in an individual displaying symptoms of anxiety, according to
the NEQ
Quality
Negative
No change
Positive
Intelligence
Attractiveness
Strength of character
Compassion
Leadership abilities
Ambition
Reliability
Mental health
17
33
53
10
69
25
28
23
62
45
26
54
9
43
47
58
2
3
2
17
3
13
6
0
(21.0)
(40.7)
(65.4)
(12.3)
(85.2)
(30.9)
(34.6)
(28.4)
(76.5)
(55.6)
(32.1)
(66.7)
(11.1)
(53.1)
(58.0)
(71.6)
(2.5)
(3.7)
(2.5)
(21.0)
(3.7)
(16.1)
(7.4)
N = 81, numbers in parentheses represent percentage of sample. NEQ = Negative Evaluation
Questionnaire.
impression of an individual's intelligence, attractiveness, compassion, ambition,
reliability, and mental health would be unchanged if they noticed that she/he was
anxious. More than 15% of participants reported that they would think an anxious
individual more compassionate and more ambitious. However, more than half
reported that they would think less of such an individual's leadership abilities and
strength of character.
In order to better understand these findings, the relationship between social
anxiety and the evaluations was examined. First, given the item distributions, it
was decided that each item except ``compassion'' and ``ambition'' would be
considered a dichotomous variable, and individuals were grouped according to
whether their evaluation would not change or would change for the worse. In the
case of ``compassion'' and ``ambition,'' individuals were grouped according to
whether the immediate impression was effected positively, negatively, or not at
all. Next, in order to determine whether evaluations on each items were
influenced by social desirability, MCSDS scores were compared across negative
change, no change, and, where applicable, positive change groups. If social
desirability played a role in the evaluations, the groups reporting no change or
positive change should have higher scores on the MCSDS. However, no
differences in MCSDS scores were observed across evaluation groups. Thus,
social desirability did not significantly influence participants' reported character
evaluations, and was not included in any further analyses.
A series of MANOVAs was then conducted comparing scores on the SIAS
and the SPS, as well as a total score on the SASS (calculated by summing all
items), across evaluation status (negative change, no change, or, where relevant,
positive change) on each item. Correlations between the three measures of social
anxiety ranged from .75 to .79, indicating that it was appropriate to include the
measures together in a MANOVA. Means and standard deviations of the SASS,
SIAS, and SPS scales across evaluation status for each item are presented on
Table 4. There were no significant differences in level of social anxiety across
52
33
26
17
28
23
10
25
Strength of
character
Attractiveness
Leadership
Intelligence
Reliability
Mental health
Compassion
Ambition
24.10
(16.69)
27.70a
(16.89)
24.92
(18.43)
30.00
(19.20)
24.07
(15.82)
27.04
(17.61)
19.80ab
(14.52)
24.92
(17.37)
42
53
57
46
61
43
44
26
n
17.15
(14.05)
17.39b
(14.28)
22.35
(15.08)
19.20
(14.25)
18.96
(14.62)
19.84
(14.95)
18.93a
(13.33)
18.38
(14.56)
13
17
na
na
na
na
na
na
n
32.47b
(20.37)
27.54
(16.06)
25
10
23
28
17
26
33
52
n
26.29a
(16.70)
27.55a
(16.59)
26.19
(18.09)
27.82
(18.55)
24.86
(14.72)
25.09
(17.06)
23.20ab
(14.13)
24.76
(16.04)
Mean
(S.D.)
42
53
57
46
61
43
44
26
n
14.54b
(8.31)
18.32b
(13.63)
21.21
(14.49)
20.48
(13.18)
20.24
(14.57)
21.49
(14.78)
19.13a
(12.97)
19.95
(14.02)
Mean
(S.D.)
No change
13
17
na
na
na
na
na
na
n
32.71b
(19.23)
26.54
(18.24)
Mean
(S.D.)
Positive
25
10
23
28
17
26
33
52
n
17.12a
(14.19)
16.39
(13.11)
18.54
(17.35)
17.41
(14.99)
17.29
(13.93)
17.65
(15.51)
14.20
(11.50)
16.88
(14.70)
Mean
(S.D.)
Negative
SPS
42
53
57
46
61
43
44
26
n
9.27b
(8.45)
13.12
(13.12)
13.35
(9.90)
12.71
(10.10)
11.65
(8.89)
13.05
(11.60)
13.04
(12.66)
13.17
(12.32)
Mean
(S.D.)
No change
13
17
na
na
na
na
na
na
n
18.65
(14.22)
13.46
(11.28)
Mean
(S.D.)
Positive
N = 81. Means sharing subscripts within each questionnaire and within each NEQ item do not differ from each other. NEQ = Negative Evaluation Questionnaire,
SASS = Symptoms of Social Scale, SIAS = Social Interaction Anxiety Scale, SPS = Social Phobia Scale. Negative, No change, and Positive refer to whether the
immediate impression of that particular quality was influenced negatively, positively, or not at all by the presence of social anxiety, based on NEQ responses. na = not
applicable (i.e., participants reported either a negative evaluation or no change in evaluation).
n
Quality
Mean
(S.D.)
Positive
Negative
Mean
(S.D.)
No change
Negative
Mean
(S.D.)
SIAS
SASS
Table 4
Means and standard deviations of social anxiety measures across evaluation groups for each item on the NEQ
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
211
212
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
groups on the qualities of ambition, leadership abilities, intelligence, reliability, or
mental health. Thus, the degree to which participants themselves experienced
social anxiety did not influence their evaluations of these qualities in other
individuals who show signs of anxiety.
However, significant differences across evaluation status were observed in the
remaining items. Anxiety measure scores were different across ``Strength of
character'' evaluation status [ F(1,36) = 4.10, P < .009]. Interpretation of univariate F tests revealed that participants who reported that they would think less of an
anxious individual's strength of character had higher scores on the SIAS
[ F(1,76) = 11.40, P < .001] and the SPS [ F(1,76) = 6.73, P < .01], indicating
higher levels of social anxiety in performance situations and those involving
interactions with others. Differences were also found across the ``Attractiveness''
ratings [ F(1,35) = 5.41, P < .002]. Examination of the univariate F tests revealed
that individuals who thought an anxious person to be less attractive had higher
scores on the SASS [ F(1,75) = 8.40, P < .005] and the SIAS [ F(1,75) = 7.16,
P < .009], suggesting that greater frequency of social anxiety symptoms and
greater anxiety when interacting with others were associated with a tendency to
evaluate others who appear anxious as being less attractive.
Finally, differences were also found on the item ``Compassion'' [ F(2,36) = 2.84,
P < .01]. Post hoc analyses were conducted comparing the three groups, and
examination of univariate analyses revealed that participants who evaluated an
anxious person as being more compassionate had higher scores on the SASS
[ F(1,77) = 10.23, P < .002] and the SIAS [ F(1,77) = 11.09, P < .001] than those
who reported no change in their view of an individual's compassion [ F(1,36) = 2.84,
P < .01], but did not have higher anxiety scale scores than those who reported a
negative evaluation. In sum, the greater the degree to which participants themselves
experienced social anxiety, the greater their tendency to report thinking less of the
strength of character and attractiveness of another individual who exhibits social
anxiety. On the other hand, higher social anxiety was associated with a tendency to
view another person who appears anxious as more compassionate.
4. Discussion
The purpose of this study was to obtain some normative data on the frequency
with which nonclinical individuals experience symptoms of social anxiety and to
examine the degree to which perceiving social anxiety in another individual
influences one's impression of various qualities of that individual. Such information is useful for correcting potentially exaggerated beliefs held by people with
social phobia regarding the social consequences of exhibiting anxiety in front of
other people. Consistent with the literature on the prevalence of shyness and
social anxiety in the general population, the data revealed that all of the anxiety
symptoms had been experienced by some of the participants at one point or
another, and that most individuals experienced symptoms of anxiety in social
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
213
situations from time to time. Thus, as hypothesized, symptoms of social anxiety
are not by any means a rare or extreme occurrence within the normal population.
As argued by Hofmann and Roth (1996), this further suggests that using a general
sample as a control group in social phobia research may result in lowered power,
given that the controls may exhibit many symptoms of anxiety. The most
frequently endorsed items on the SASS tended also to be the ones that socially
anxious individuals are particularly concerned about. That is, the most frequently
endorsed symptoms were ones that are externally detectable, such as blushing,
sweating, laughing, or smiling inappropriately, and having difficulty expressing
oneself. These data might be useful as compelling evidence that socially anxious
individuals may underestimate the extent to which others become anxious and
actually exhibit visible signs of anxiety.
The results of this study also revealed that the vast majority of participants
reported that if they were to notice that someone was anxious, it would not
influence their perception of that individual's intelligence, ambition, reliability, or
mental health. Given that these evaluations were unrelated to scores on the
MCSDS, we can conclude that this is not simply a result of a prosocial bias in
responding. At the same time, however, individuals did report that they would
think less of an individual's leadership abilities if that individual exhibited
anxiety, and a strong majority reported that they would think less of that
individual's strength of character. This finding could reflect a general negative
view about anxiety, or it could reflect an idealistic view of what strong leaders
should be like. Regardless, it suggests that therapists might need to be cautious
when challenging beliefs about society's perception of anxiety, and be willing to
recognize that negative evaluation is possible, particularly in the absence of any
other information about the individual (although the therapist would, then, want
to address beliefs about the importance of being evaluated negatively, challenging
the position that this is truly catastrophic as well as the position that immediate
impressions are not amenable to change in the presence of other kinds of
information about the individual).
The study also revealed that participants who themselves were socially
anxious were likely to view others who show signs of anxiety as less attractive
and as having less strength of character. Ironically, individuals high in social
anxiety also reported that other individuals exhibiting signs of anxiety would be
more compassionate. These data suggest that individuals higher in social anxiety
may, in fact, be somewhat less compassionate. This contradicted the hypothesis
that individuals higher in social anxiety would not apply the same stringent
standards for interpersonal behavior to others as they do to themselves. These
data may reflect more general negative beliefs among individuals with social
anxiety about what it means to be anxious. Since individuals with social phobia
may be more likely to evaluate others negatively, they may then have an
exaggerated sense of the extent to which signs of their own anxiety are evaluated
negatively by others, at least nonanxious others. This may reflect more general
beliefs about what anxiety is and why it is experienced that are both erroneous
214
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
and negative (e.g., ``Anxiety is a sign of mental weakness,'' ``Anxiety is an
inappropriate emotion'') that could be targeted in treatment. Offering socially
anxious individuals some normative information on the extent to which people in
general experience anxiety could help alleviate the concern that it is unusual to
experience anxiety and that only the ``weak'' ever experience it.
The results of this study must be considered preliminary given the sample size
and that the data were based on subjective self-report about a hypothetical
situation from a nonclinical sample. However, the data do suggest that further
investigation of these questions is worthwhile. It would be interesting to examine
the impact of anxiety on immediate impressions in an experimental paradigm in
which the anxiety level of the individual being evaluated is manipulated,
comparing results across nonclinical and clinical social phobic groups. Nonetheless, the data do suggest that social anxiety is experienced by most people
from time to time, and that there may be prevalent negative attitudes towards
anxiety within the general population and within socially anxious individuals in
particular. These data may be useful in cognitive restructuring work with socially
anxious individuals.
References
Alden, L. E., & Wallace, S. T. (1995). Social phobia and social appraisal in successful and unsuccessful social interactions. Behaviour Research and Therapy, 33, 497 ± 506.
American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders
(4th ed.). Washington, DC: American Psychiatric Association.
Brown, C. J., Turovsky, J., Heimberg, R. G., Juster, H. R., & Barlow, D. H. (1997). Validation of the
social interaction anxiety scale and the social phobia scale across the anxiety disorders. Psychological Assessment, 9, 21 ± 27.
Bruch, M. A., Borsky, J. M., Collins, T. M., & Berger, P. A. (1989). Shyness and sociability reexamined: a multicomponent analysis. Journal of Personality and Social Psychology, 57, 904 ± 915.
Clark, D. M., & Wells, A. (1995). A cognitive model of social phobia. In: R. G. Heimberg, M.
R. Liebowitz, D. A. Hope, & F. R. Schneier (Eds.), Social phobia: diagnosis, assessment and
treatment ( pp. 69 ± 93). New York: Guilford.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirability independent of psychopathology. Journal of Consulting Psychology, 24, 349 ± 354.
Ernulf, K. E., & Innala, S. M. (1998). Masculinity, femininity and sexual attractiveness: a crossnational perspective. Scandinavian Journal of Sexology, 1, 101 ± 120.
Hofmann, S. G., & Roth, W. T. (1996). Issues related to social anxiety among controls in social phobia
research. Behavior Therapy, 27, 79 ± 91.
Mattick, R. P., & Clarke, J. C. (1998). Development and validation of measures of social phobia
scrutiny fear and social interaction anxiety. Behaviour Research and Therapy, 36, 455 ± 470.
Mattick, R. P., Page, A., & Lampe, L. (1995). Cognitive and behavioural aspects. In: M. B. Stein (Ed.),
Social phobia: clinical and research perspectives. Washington, DC: American Psychiatric Press,
189 ± 228.
McEwan, K. L., & Devins, G. M. (1983). Is increased arousal in social anxiety noticed by others?
Journal of Abnormal Psychology, 92, 417 ± 421.
Millham, J., & Jacobsen, L. I. (1978). The need for approval. In: H. London, & J. E. Exner (Eds.),
Dimensions of personality. New York: Wiley.
C. Purdon et al. / Anxiety Disorders 15 (2001) 203±215
215
Rapee, R. M., & Heimberg, R. G. (1997). A cognitive ± behavioral model of anxiety in social phobia.
Behaviour Research and Therapy, 35, 741 ± 756.
Rubin, Z. (1970). Measurement of romantic love. Journal of Personality and Social Psychology, 16,
265 ± 273.
Scheling, A., & Emmelkamp, P. M. G. (1993). Cognitive and behavioural treatments of fear of
blushing, sweating or trembling. Behaviour Research and Therapy, 31, 155 ± 170.
Scholing, A., Emmelkamp, P. M. G., & Van Oppen, P. (1996). Cognitive ± behavioral treatment of
social phobia. In: V. B. Van Hasselt, & M. Hersen (Eds.), Sourcebook of psychological treatment
manuals for adult disorders ( pp. 123 ± 177). New York: Plenum.
Stein, M. B., Walker, J. R., & Forde, D. R. (1994). Setting diagnostic thresholds for social phobia:
considerations from a community survey of social anxiety. American Journal of Psychiatry, 151,
408 ± 412.
Turner, S. M., Beidel, D. C., & Townsley, R. M. (1990). Social phobia: relationship to shyness.
Behaviour Research and Therapy, 28, 497 ± 505.
Wells, A. (1997). Cognitive therapy of anxiety disorders: a practice manual and conceptual guide.
New York: Wiley.
Zimbardo, P. G. (1977). Shyness: what it is, what to do about it. Reading: Addison-Wesley.
View publication stats