- Email: [email protected]
Social phobia and perfectionism
Person. individ. Difj: Vol. 21. No. 3, pp. 403410, 1996 Copyright c 1996 Elsevier Science Ltd Printed in Great Britain. All rights reserved 0191~8869/96 $15.00+0.00
Pergamon
SOCIAL PHOBIA AND PERFECTIONISM Harlan R. Juster,‘* Richard G. Heimberg,’ Randy 0. Frost,’ Craig S. Holt3 Jill I. Mattia’ and Karen Faccenda’ ‘Social Phobia Program, Center for Stress and Anxiety Disorders, Department of Psychology, University of Albany, State University of New York, Albany, NY, U.S.A., *Department of Psychology, Smith College, Northampton, MA, U.S.A. and ‘University of Iowa and Iowa City Veterans Administration Medical Center, Iowa City, IA, U.S.A. (Received 4 September 1995)
Summary-Clinical evidence suggests that components of perfectionism may have special relevance to social phobia. This study examines this relationship by comparing 61 patients with social phobia and 39 community volunteers with no anxiety disorder on Frost, Marten, Lahart and Rosenblate’s (1990) Multidimensional Perfectionism Scale (MPS). Social phobia patients scored higher on subscales assessing concern over mistakes, doubts about actions, and perceived parental criticism. Community volunteers scored higher on the organization subscale. It was further hypothesized that, for the social phobia patients, perfectionism would be associated with greater symptom severity. Correlational analysis confirmed that the Concern over Mistakes and Doubts about Actions subscales of the MPS were consistently associated with greater social anxiety, trait anxiety, and general psychopathology. Implications for the treatment of social phobia patients are considered. Copyright 0 1996 Elsevier Science Ltd.
INTRODUCTION
Social phobia is a fear of situations in which an individual may be under the scrutiny of others and expects to act in an embarrassing or humiliating manner (DSM-IV; American Psychiatric Association, 1994). Individuals with social phobia often fear making unintelligent, unintelligible or socially inappropriate comments. They may fear that anxiety symptoms, such as blushing, sweating, or shaking, will be obvious to others, thereby ‘giving away’ that they are anxious. These and other fears about being the center of attention are often associated with doubt about one’s competence. Although we have learned much about social phobia in the last several years, we know little about personality dimensions which may influence the severity of patients’ distress or have implications for how they respond to various treatment strategies. This study examines one such dimension, perfectionism. Frost, Marten, Lahart and Rosenblate (1990) define perfectionism as the setting of excessively high personal standards plus a tendency to be overly self-critical upon failure to meet those standards. Frost et al. (1990) developed the Multidimensional Perfectionism Scale (MPS) to measure the key concepts of excessive personal standards and overconcern about making mistakes, and included other scales which have conceptual relevance to perfectionism including doubts about the quality of one’s actions, parental expectations, parental criticisms, and extreme organization. Perfectionism has been implicated as a factor in a variety of mental disorders including obsessivecompulsive disorder (e.g. McFall & Wollersheim, 1979), eating disorders (e.g., Garner, Garfinkel & Bemis, 1982) depression (Burns, 1980), and anxiety (e.g., Deffenbacher, Zwemer, Whisman, Hill & Sloan, 1986) and appears to have relevance to the clinical manifestations of social phobia. For example, our clinical experience suggests that individuals with social phobia often set virtually unattainable goals for themselves (excessive standards of performance) in feared social situations, thereby guaranteeing the occurrence of ‘mistakes’ in these situations. They also appear to make unreasonable comparisons when judging their own competence. Their public speaking efforts may be compared with the performance of television commentators or public officials while their social interactional behaviors are compared with those of the ‘life of the party’. When social phobic individuals do perform up to ‘par’, they are clearly not objective observers of their own behavior
*To whom all correspondence should be addressed at: Center for Stress and Anxiety Disorders, Pine West Plaza Building 4, Washington Avenue Extension, Albany, NY 12205, U.S.A.
404
Harlan R. Juster et af.
and will often devalue their otherwise solid performance or disqualify the situation as one that is atypical in some critical way (e.g. “I did well because the audience felt sorry for me” or “he was trying to be nice”) (Rapee & Lim, 1992). Furthermore, social phobic individuals engage in frequent avoidance of unfamiliar activities, possibly for fear of performing below standards. Heimberg and his colleagues (Heimberg & Becker, in press; Heimberg, Juster, Hope & Mattia, 1995) have developed a model of social phobia which may help explain the link between social phobia and perfectionism. Briefly, this mode1 proposes that a genetic susceptibility combines with early experience to sensitize individuals to view social encounters as threatening experiences. A set of (perfectionistic) beliefs develops which primes individuals to expect negative consequences in social situations, resulting in anxiety and often avoidance. Specifically, individuals with social phobia come to believe that (1) social situations are potentially dangerous (i.e. humiliating), (2) danger may be averted only by social performance above a very high standard (i.e. perfect performance), and (3) they are not capable of achieving the necessary standard to avoid danger. Achieving a certain standard may be hampered by over attention to mistakes in social situations. A vicious cycle may exist in which the expectation of mistakes produces excessive focus on evaluation of one’s performance which, in turn, causes a reduction in attention to the actual task, making mistakes more likely. Any mistake reduces the likelihood of attaining the desired standard, and therefore becomes an indicator of failure. Even when mistakes are within normal limits of magnitude or frequency for a particular situation, over attention to them may result in a biased view of the situation as a failure. Thus, excessive standards and concern over mistakes appear to be closely linked characteristics of individuals with social phobia. Although there is no direct evidence linking perfectionism and social phobia, several studies have examined the relationship among perfectionism, excessive standard setting, and social or performance anxiety in nonclinical populations. Each of these areas will be briefly considered. Frost and Henderson (1991) found that, for a sample of female college athletes, perfectionistic concern over mistakes was positively correlated with anxiety about performance in competition. Conversely, concern with mistakes and self-doubts were negatively correlated with self-confidence regarding athletic performance. Frost and Marten (1990) showed that perfectionistic college students, compared with nonperfectionistic students, experienced higher levels of negative affect while completing a writing task under conditions of high evaluative threat, a situation analogous to social encounters for social phobia patients. High perfectionistic students reported that doing well on the writing task was more important to them than it was to low perfectionistic students and believed they should have done better on the task as well. Test-taking is another area of academic performance which may activate evaluative anxiety. In fact, DSM-IV describes test anxiety as a variant of social phobia in which individuals fear indirect evaluation. Juster, Brown, Heimberg, Makris, Leung and Frost (1994) examined the association between Frost’s MPS and a measure of test anxiety in a sample of undergraduates. Predictably, the Concern over Mistakes and Doubts about Actions subscales were positively correlated with the cognitive and physiological responses associated with test anxiety. Mor, Day, Flett and Hewitt (1995) found that performance anxiety in professional musicians, actors, and dancers was positively correlated with perfectionistic personal and social standards. The largest association was found between socially prescribed perfectionism (i.e. standards based on others’ expectations) and anxiety. Clearly, perfectionism is associated with anxiety and other negative reactions in situations where individuals believe they are being evaluated. Negative reactions associated with concern over mistakes have received particular attention. This component of perfectionism is defined as negative cognitive reactions to mistakes, the tendency to equate mistakes with failure, and the expectation of negative responses from others regarding mistakes. Athletes high on concern over mistakes reported that they were more likely to focus on the negative aspects of their mistakes (Frost & Henderson, 1991). These include increased worry about others’ reactions, greater likelihood of feeling they let themselves down, more pressure to make up for the mistake, greater difficulty forgetting about it, and more recurrent images of the mistake. In another study, students scoring high on the Concern over Mistakes subscale of the MPS reported more negative affect and lower self-confidence, believed others judged them more harshly, and were more likely to conceal the details of their performance than low Concern over Mistakes
Social phobia and perfectionism
405
students after completing a task which elicited a high frequency of mistakes (Frost, Turcotte, Heimberg, Mattia, Holt & Hope 1995). Furthermore, high Concern over Mistakes students were more likely to believe that they ‘should’ have performed to a higher standard than their low-Concern over Mistakes counterparts. In contrast to the evidence regarding Concern over Mistakes, the Personal Standards subscale has not been associated with performance anxiety and the other negative reactions noted above, yet this construct has figured prominently in the definition of perfectionism and in our model of social phobia. In fact, Personal Standards has been associated with more positive strivings, as has been suggested by factor analysis (Frost, Heimberg, Holt, Mattia & Neubauer, 1993b). Alden, Bieling and Wallace (1994) noted that standards which exceed ability may be a motivating factor if the discrepancy between standards and ability is not too great. Thus, the effect of standards in evaluative situations may be complex. For example, Juster et al. (1994) found that Personal Standards interacted with the worry (cognitive) component of test anxiety to predict academic performance. Students scoring low on the worry component attained course grades directly related to their level of personal standards. High standards were associated with high academic performance and low standards were associated with poor performance. However, the academic performance of students scoring high on worry was unrelated to personal standards. Alden et al. (1994) showed that socially anxious undergraduates set personal standards in line with their abilities but lower than nonanxious students. Importantly, socially anxious students believed their abilities fell short of the standards set by others. Thus, personal standards may be lower in individuals with social phobia but still within their expected capabilities. However, they may believe their abilities fall short of the expectations that others hold for them. This is consistent with the clinical observation that social phobics can often perform tasks when alone without fear but are distressed by the same behavior when in the presence of others. The current study is the first investigation of the relationship of the dimensions of perfectionism and social phobia. We compared the MPS subscale responses of a group of social phobic subjects with those of a group of non-social phobic community volunteers. We predicted that social phobic subjects would score higher on this measure of perfectionism, most specifically the Concern over Mistakes subscale. For the social phobic patients only, the relationship of MPS subscale scores to measures relevant to social phobia and general psychopathology was examined. We predict that greater reliance on a perfectionistic style would be associated with greater severity of social phobia and that this relationship would not be better accounted for by patients’ score on a measure of depression.
METHOD Subjects Social phobic subjects. Sixty-one patients (38 males, 23 females) meeting DSM-III-R (American Psychiatric Association, 1987) criteria for social phobia participated in this study. Social phobic patients requesting treatment were identified by the Anxiety Disorders Interview Schedule-Revised (ADIS-R; DiNardo & Barlow, 1988). Subjects were rated on the O-8 Clinician’s Severity Rating (CSR), included in the ADIS-R, and only subjects who received a rating of four or greater (indicative of moderate impairment and need for treatment) were included in the study. The ADIS-R has been shown to have high rates of inter-rater agreement when social phobia is assigned as the principal diagnosis (k = 0.79; DiNardo, Moras, Barlow, Rapee & Brown, 1993). Patients were excluded from the study if they received a comorbid diagnosis of current major depression, bipolar disorder, psychotic disorder or active drug or alcohol dependence within the past three months. Participation in the study was part of the pretreatment assessment administered to all patients who enter treatment at the Social Phobia Program of the Center for Stress and Anxiety Disorders, University at Albany, State University of New York. Community volunteers. A nondisordered control sample of 2 1 women and 18 men from the Albany area were recruited by response to bulletin board advertisements requesting the assistance of persons who were neither anxious nor depressed. A detailed telephone screening interview was conducted with each respondent to rule out the presence of an anxiety disorder, affective disorder, or substance
406
Harlan R. Juster et al. Table I. Demographic characteristics for social phobia patients and community volunteers Social phobia patients (N=61) Mean age % Male % Married % Employed full-time % College graduate
36.6(9.6) 62 41 62 70
Community volunteers (N = 39) 34. I (10.0) 46 49 82 46
Figures in parentheses are standard deviations.
abuse or treatment for any of these conditions within the past two years. Control subjects were included in the order they passed screening, with the constraints that the sample be balanced for gender and equated for age with the social phobic sample. Demographic characteristics of the social phobic and community volunteer groups are summarized in Table 1. Measures The Multidimensional Perfectionism Scale (MPS; Frost et al., 1990) is a 35item self-report measure that contains six subscales related to the construct of perfectionism: (1) concern over mistakes (CM); (2) doubts about actions (DA); (3) personal standards (PS); (4) parental expectations (PE); (5) parental criticism (PC); and (6) organization (OR). The CM subscale is a measure of overly critical self-evaluation, a key concept in pathological perfectionism according to Frost et al. (1990). This scale includes items such as “People will probably think less of me if I make a mistake” and “If I fail at work/school, I am a failure as a person”. Another measure of critical self-evaluation, the DA subscale reflects a general dissatisfaction with or uncertainty about the quality of one’s effort or whether one has chosen the right course of action. Items in this subscale include “Even when I do something very carefully, I often feel that it is not quite right” and “It takes me a long time to do something right”. Of the MPS subscales, CM and DA have been most consistently related to measures of self-critical depression, compulsivity, procrastination and psychopathology (Frost et al., 1990). The PS subscale reflects a traditional view of perfectionism as the setting of excessively high standards and appears to have both positive and negative aspects (Frost et al., 1990). This is reflected in scale items such as “It is important for me to be thoroughly competent in everything I do” and “I expect higher performance in my daily tasks than most people”. Expectations and critical evaluations of parents towards their children have a role in perfectionism and may be etiologically associated with personal expectations and self-evaluations (Frost, Lahart & Rosenblate, 1991). An example of an item from each of these scales is “My parents wanted me to be the best at everything” (PE) and “My parents never tried to understand my mistakes” (PC). The OR subscale reflects an individual’s tendency to emphasize orderliness and precision in daily tasks. OR scale items include “Organization is very important to me” and “I am a neat person”. Reliability and validity of the MPS. Internal consistency has ranged from good to excellent for each of the subscales (US= 0.77-0.93) and the total perfectionism score (c( = 0.90; Frost et al., 1990). The total perfectionism score (equal to the sum of all subscales except OR) was significantly correlated with other measures of perfectionism while CM and DA have been most consistently correlated with measures of general psychopathology (Frost et al., 1990). CM was directly related to procrastination while PS and OR were correlated with lower frequencies of procrastination, supporting speculation about the positive aspects of these two scales. As noted previously, MPS scores have predicted subjects’ reactions to evaluative threat (Frost & Marten, 1990) and to making mistakes (Frost et al., 1995). Other assessments. Instruments measuring various aspects of social and general anxiety and general psychopathology were administered to the social phobic patients. The Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1989) assesses anxiety in social interaction situations while the Social Phobia Scale (SPS; Mattick & Clarke, 1989) focuses primarily on anxiety while being observed. The Social Phobia subscale of the Fear Questionnaire (FQSO; Marks & Mathews, 1979)
Social phobia and perfectionism
407
provides a measure of avoidance of typical social situations and the Fear of Negative Evaluation Scale (FNE; Watson & Friend, 1969) addresses patient’s concerns about the disapproval of others. Also administered were the Trait form of the State-Trait Anxiety Inventory (STAI-T; Spielberger, Gorsuch, Lushene, Vagg & Jacobs, 1983) the Global Symptom Index (GSI) of the Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1977), and the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock & Erbaugh, 1961). The interviewer-administered CSR and the Fear (LSAS-F) and Avoidance (LSAS-A) Subscales of the Liebowitz (1987) Social Anxiety Scale were included as interviewer-rated measures of social anxiety.
RESULTS Demographic characteristics
Social phobic patients were compared with community volunteers on several demographic variables. There were no significant differences between the two groups on gender, age, or marital status. They differed on educational level, such that a greater percentage of social phobic patients had graduated from college (Yates corrected x2 (1, N = 100) = 4.94, P = 0.026). Employment level was examined using a number of different groupings. When the number of individuals employed fulltime was compared with all other categories combined (part-time, student, homemaker, and unemployed), there was a trend for a greater percentage of community volunteers to be employed on a full-time basis (Yates corrected x2 (1, N = 100) = 3.53, P = 0.06). There were no differences when full- and part-time employment were combined and compared with the remaining groups. A series of analyses of variance revealed that neither educational level nor employment status was related to any of the MPS subscales, so these variables are not considered further in the analyses of our findings. Furthermore, males and females with social phobia did not differ on any MPS subscales, a finding consistent with Frost, Heimberg, Holt and Mattia’s (1993a) results comparing a large sample of undergraduates. Male and female social phobics also showed no differences on any measures of social anxiety or general psychopathology. All subsequent analyses combine these two groups. MPS subscales
Differences between the patients and community volunteers on MPS subscale scores were evaluated by multivariate analysis of variance. With the six subscales scores as the dependent variables, the multivariate test was significant (Wilks’ i = 0.793; F(4,93) = 4.03, P < 0.01). Follow-up univariate analyses of variance indicated that social phobics and community controls differed on CM, DA, PC, and OR but not on PS or PE. Means, standard deviations, and results of univariate F-tests are provided in Table 2. For CM, DA, and PC, social phobia patients scored higher than community controls. The community controls scored higher than the social phobics on OR.
Table 2. Means, standard deviations, univariate F-ratios, and probability values for Multidimensional Perfectionism Scale scores and differences between groups
MPS subscale CM PS DA PC PE OR
Social phobia patients (N = 61)
Community volunteers (N = 39)
F
P
25.6(7.5) 23.1(5.8) 10.9(3.7) 9.3 (3.9) 12.7(6.0) 20.1(5.4)
20.3(6.7) 23.4(4.9) 9.0(3.5) 7.8(3.1) 12.5(3.9) 23.1(5.0)
12.83 0.07 7.21 4.15 0.06 7.67
0.001 o.“do9 0.044 o.noso7
MPS = Multidimensional Perfectionism Scale; CM = Concern Over Mistakes; PS = Personal Standards; DA = Doubts About Actions; PC = Parental Criticism; PE = Parental Expectations; OR = Organization. Figures in parentheses are standard deviations.
408
Harlan R. Juster ef al.
Table 3. Zero-order and semi-partial correlations between MPS subscales and measures of social and general anxiety for social phobia patients
CM
PS
MPS subscales DA
PE
PC
OR
0.25 0. I5 0.07 0.05
-0.05 -0.08 0.22 0.00
Social an.riet.v Se@eport
Fear questionnaire-Social Phobia Social interaction anxiety scale Social phobia scale Fear of negative evaluation scale
0.46’ 0.36 0.26 0.50;
(0.38t)
(0.39t)
-0.02 -0.03 0.08 0.00
0.44* 0.52’ 0.36 0.43’
(0.30) (0.33t) (0.27)
0.12 0.16 0.04 0.16
-0.05 0.08 0.03
0.39’ 0.531 0.50*
(0.28) (0.37t) (0.35t)
0.07 0.23 0.20
0.06 0.26 0.16
0.02 -0.03 0.05
0.09 0.1 I 0.19
0.37 0.61’ 0.47*
(0.37t) (0.21)
0.07 0.07 0.09
0.16 0.24 0.21
-0.06 0.02 0.07
Intemiewer-administered
ADIS-R clinician’s severity rating Liebowitz social anxiety scale-fear Liebowitz social anxiety scale-avoidance
0.21 0.44* 0.36
General psvchopathology Beck depression inventory State-trait anxiety inventory-trait SCL-90-R global symptom index
0.24 0.48* 0.40’
(0.33t)
(0.3lt) (0.23t)
MPS = Multidimensional Perfectionism Scale; CM = Concern Over Mistakes; PS = Personal Standards; DA = Doubts About Actions; PC = Parental Criticism; PE = Parental Expectations; OR = Organization; SCL-90-R = Symptom Checklist 90-Revised; ADISR = Anxiety Disorders Interview Schedule-Revised Values in parentheses are semi-partial correlations controlling for Beck Depression Inventory score. For zero-order correlations, * Indicates P < 0.002 for self-report social anxiety measures and P < 0.003 for all other measures. For semi-partial correlations. t indicates P < 0.01 for CM and P -c 0.0063 for DA.
MPS, social anxiety, andpsychopathology
The relationship between perfectionism, social anxiety, and general psychopathology was examined for the social phobic patients. Correlations between MPS subscales and other measures were calculated. Measures were classified as self-reports of social anxiety, interviewer assessments of social anxiety, or self-reports of general psychopathology. Given the large number of zero-order correlations, Bonferroni corrections were applied to each group of measures. Thus, for the four self-report measures of social anxiety, probabilities of 0.002 or less were considered significant. For interviewer-administered measures of social anxiety and self-report measures of general psychopathology (three measures each), probability values of 0.003 or less were considered significant. Correlations are provided in Table 3. Several significant correlations were found between CM and DA and measures of social anxiety and psychopathology. CM was significantly correlated with FQSO, FNE, STAI-T, GSI, and the LSAS-F. DA was significantly correlated with all measures except the SPS and the BDI. No significant zero-order correlations were found between PS, PE, PC, OR subscales and any other measure. Significant zero-order correlations were followed by examination of semi-partial correlations controlling for the effect of depression (BDI). Semi-partial correlations were examined since depression is correlated with most measures of anxiety and our interest was in the unique relationship between dimensions of perfectionism and social anxiety. Once again, Bonferroni corrections were applied. For CM, semi-partial correlations with a probability of 0.01 or less were considered significant, whereas for DA, probabilities of less than 0.0063 were considered significant. All correlations involving CM and four of eight correlations with DA remained significant after controlling for depression. These findings generally confirm the pattern of relations between CM, DA, and measures of social anxiety and general psychopathology, although the findings for DA are somewhat less robust than suggested by the zero-order correlations.* *The helpful comments of an anonymous reviewer to an earlier version of this manuscript suggest an alternati-/e strategy. First, a single composite factor score was developed composed of the self-report and interviewer-generated social anxiety measures. The alpha reliability for this composite of seven measures was 0.88. Hierarchical multiple regression was performed on the composite score which was regressed on BDI and SCLGSI entered as the first block of the regression equation (TRAIT was not included in this block as it would have reduced the variance available for prediction to practically nil). The second block (the one of interest in this analysis) was composed of the CM, DO, and PS subscales of the MPS. The second block produced a significant increment in variance accounted for over the first block (Fchange = 3.95, P < 0.02) indicating that the linear combination of CM, DO, and PS accounted for a significant proportion of variance in the composite social anxiety measure beyond that predicted by a measure of depression and general psychopathology. Examination of the semi-partial correlations for the MPS subscales (0.16, 0.17, and - 0.17 for CM, DO, and PS, respectively) indicate virtually equivalent contributions of these predictors although all three failed to reach significance independently (P = 0.08) despite their significant joint contribution.
Social phobia and perfectionism
409
DISCUSSION Patients with social phobia were more likely to achieve extreme scores on a measure of perfectionism than a sample of community volunteers. Social phobics were more concerned about making mistakes and more likely to doubt their own actions; they reported more parental admonishment as a result of not living up to parent’s standards and less concern about orderliness and organization than community volunteers without an anxiety disorder. These empirical findings support our clinical observations that social phobic patients expect to make critical errors and doubt their abilities in social interaction and public performance. They tend to focus more attention on their mistakes, overvalue the importance of mistakes, and blame themselves for poor outcomes. Social phobic patients scored higher on the Concern over Mistakes subscale than the community volunteers. Concern over Mistakes is associated with increased negative affect, the interpretation of mistakes as a sign of specific and global failure, concomitant reduction in motivation to persist at an activity, and the assumption that others also see them as a failure. Thus, the stage is set for the cycle of fear and avoidance of social situations which characterizes social phobia. Social phobic individuals also appear to devalue their performance and experience much self-doubt (as indicated by their higher scores on the Doubts about Actions subscale) when they have chosen a particular course of action. Self-doubt may focus on one’s ability to perform required social behaviors, in giving the ‘correct’ response or asking the ‘proper’ questions in a social interaction, or in meeting personal goals. Consistent with the model described earlier (Heimberg et al., 1995). social anxiety occurs when individuals believe they will be unable to create a desired impression in a social situation (Leary & Atherton, 1986; Schlenker 8c Leary, 1982). For the social phobic patients, greater concern over mistakes and doubts about actions were indicative of more severe social anxiety, trait anxiety, and general psychopathology. The associations remained even after the effect of depression was statistically controlled. Notably, Doubts about Actions and Concern over Mistakes were also correlated with interviewer-rated measures. The finding of an association between components of perfectionism and social anxiety z:cross different modes of assessment lends further credence to this relationship. Although social phobics reported more parental criticism than community volunteers, they reported similar levels of parental expectations. Conceptually, parental criticism appears to be more closely associated with other negative aspects of perfectionism, especially concern over mistakes. Expectations, however, appear to be more highly associated with personal standards, a measure of self-expectations. This pattern of relationships is supported by interscale correlations from a previous study (Frost et al., 1990). Parental criticism may have etiological significance for social phobia. Several studies retrospectively examined the parental-rearing styles in the families of social phobic patients (Arrindell, Emmelkamp, Monsma & Brilman, 1983; Parker, 1979). Bruch, Heimberg, Berger and Collins (1989) showed that social phobic patients perceived their parents as placing greater emphasis on the opinions of others regarding appropriate behavior than agoraphobic patients. Bruch and Heimberg (1994) replicated this finding and further demonstrated that parents of social phobics relied more on the use of shame as a disciplinary technique than other parents. They further argue that this pattern of repeatedly sensitizing children to the scrutiny of others is a key element in etiological mechanisms of social anxiety (Buss, 1980). The two groups did not differ on their level of personal standards, a finding consistent with previous work by Frost and his colleagues. This dimension of perfectionism has both positive and negative elements and may be common both among individuals who set and successfully strive for high goals and those who set goals beyond their capacity (Frost et ul., 1993b). As noted earlier, the distinguishing feature may be the magnitude of the discrepancy between standards and perceived ability (Alden et al., 1994). A limitation of this study is that the only comparison group consisted of nondisordered community volunteers. Thus, although the effects of depression were statistically controlled, it is not clear how the current pattern of findings might have differed had an alternative group of patients, for example with panic disorder, been included for comparison. Future research might also examine whether differences in patients’ perfectionistic styles are predictive of treatment outcome and whether treatment can alter those styles.
410
Harlan R. Juster et al.
Acknowledgements --Preparation
of this article was supported in part by grant 44119 to Richard G. Heimberg from the National Institute of Mental Health. Portions of this paper were presented at the annual meeting of the Association for Advancement of Behavior Therapy, Boston, MA, November 1992. Jill Mattia is now at the Deiartment of Psychiatry, Brown Medical School at Rhode Island Hospital, Providence, Rhode Island.
REFERENCES Alden, L. E., Bieling, P. J. & Wallace, S. T. (1994). Perfectionism in an interpersonal context: A self-regulation analysis of dysphoria and social anxiety. Cognitive Therapy and Research, 18, 291-316. American Psychiatric Association (1987).Diagnostic and Statistical Manual of Mental Disorders (3rd rev. edn). Washington, DC: Author. American Psychiatric Association (1994).Diagnostic and Statistical Manual of Mental Disorders (4th edn). Washington, DC: Author. Arrindell, W. A., Emmelkamp, P. M. G., Monsma, A. & Brilman, E. (1983). The role of perceived parental rearing practices in the etiology of phobic disorders: A controlled study. British Journal of Psychiatry, 143, 183-187. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-511.
Bruch, M. A. & Heimberg, R. G. (1994). Differences in perceptions of parental and personal characteristics between generalized and nongeneralized social phobics. Journal of Anxiety Disorders, 8, 155-168. Bruch, M. A., Heimberg, R. G., Berger, P. & Collins, T. M. (1989). Social phobia and perceptions of early parental and personal characteristics. Anxiety Research, 2, 5765. Burns, D. (1980). The perfectionist’s script for self-defeat. Psychology Today, November, 34-51. Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco: Freeman. Deffenbacher, J. L., Zwemer, W. A., Whisman, M. A., Hill, R. A. & Sloan, R. D. (1986). Irrational beliefs and anxiety. Cognititre Therapy and Research, IO, 281-292.
Derogatis, L. R. (1977). SCL-90 administration, scoring and procedures manual for the revised version. Baltimore: Johns Hopkins University School of Medicine. DiNardo, P. A. & Barlow, D. H. (1988). The Anxiety Disorders Interuiew Schedule-Revised(ADIS-R). Albany, NY: Graywind. DiNardo, P. A., Moras, K., Barlow, D. H., Rapee, R. M. & Brown, A. A. (1993). Reliability of DSM-III-R anxiety disorder categories: Using the Anxiety Disorders Interview Schedule-Revised. Archives of General Psychiatry, 50, 251-256. Frost, R. 0.. Heimberg, R. G., Holt, C. S. & Mattia, J. I. (1993a). Perfectionism andgender. Unpublished manuscript. Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I. & Neubauer, A. L. (1993b). A comparison of two measures of perfectionism. Personality and Individual Differences, 14, 119-126. Frost. R. 0. & Henderson, K. J. (1991). Perfectionism and reactions to athletic competition. Journal of Sport and Exercise Psycholog.y, 13, 323-335.
Frost, R. O., Lahart, C. M. & Rosenblate, R. (1991). The development of perfectionism. Cognitive Therapy and Research, IS, 469490.
Frost, R. 0. & Marten, P. A. (1990). Perfectionism and evaluative threat. Cognitive Therapy and Research, 14, 5599572. Frost, R. O., Marten, P. A., Lahart, C. & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitiue Therapy and Research, 14,449-468.
Frost, R. O., Turcotte, T. A., Heimberg, R. G., Mattia, J. I., Holt, C. S. & Hope, D. A. (1995). Reactions to mistakes among subjects high and low in perfectionistic concern over mistakes. Cognitive Therapy and Research, 19, 195-205. Garner, D., Garfinkel, P. & Bemis, K. (1982). A multidimensional psychotherapy for anorexia nervosa. International Journal of Eating Disorders, I, 3-64.
Heimberg, R. G. & Becker, R. E. (in press). Treatment of socialfears andphobias. New York: Guilford. Heimberg, R. G., Juster, H. R., Hope, D. A. & Mattia, J. I. (1995). Cognitive-behavioral group treatment for social phobia: Description, case presentation, and empirical support. In Stein, M. (Ed.) Socialphobia: Clinical and research perspectives (pp. 293-321). American Psychiatric Press. Juster, H. R., Brown, E. J., Heimberg, R. G., Makris, G. S., Leung, A. W. & Frost, R. 0. (1994). Effect of test anxiety and perfectionism on academic performance. Poster presented at the annual meeting of the Association for Advancement of Behavior Therapy, San Diego, CA. Leary. M. R. & Atherton, S. C. (1986). Self-efficacy, social anxiety, and inhibition in interpersonal encounters. Journal of Social and Clinical Psychology, 4, 256267.
Liebowitz, M. R. (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22, 141-173. Marks, I. M. & Mathews, A. M. (1979). Brief standard self-rating for phobic patients. Behaviour Research and Therapy, 17, 263-267.
Mattick, R. P. & Clarke, J. C. (1989). Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Unpublished manuscript. McFall, M. E. & Wollersheim, J. P. (1979). Obsessive-compulsive neurosis: A cognitive-behavioral formulation and approach to treatment. Cognitive Therapy and Research, 3, 333-348. Mor, S., Day, H. I., Flett, G. L. & Hewitt, P. L. (1995). Perfectionism, control and components of performance anxiety in professional artists. Cognitive Therapy and Research, 19, 207-225. Parker, G. (1979). Reported parental characteristics of agoraphobics and social phobic% British Journal of Psychiatry, 135, 555-560.
Rapee, R. M. & Lim, L. (1992). Discrepancy between self- and observer ratings of performance in social phobics. Journal of Abnormal Psychology, 101, 727-731.
Schlenker, B. S. & Leary, M. R. (1982). Social anxiety and self-presentation: A conceptualization and model. Psychological Bulletin, 92, 64-669.
Spielberger, C., Gorsuch, R., Lushene, R., Vagg, P. & Jacobs, G. (1983). Manualfor the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. Watson, D. & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33,448-457.
Pergamon
SOCIAL PHOBIA AND PERFECTIONISM Harlan R. Juster,‘* Richard G. Heimberg,’ Randy 0. Frost,’ Craig S. Holt3 Jill I. Mattia’ and Karen Faccenda’ ‘Social Phobia Program, Center for Stress and Anxiety Disorders, Department of Psychology, University of Albany, State University of New York, Albany, NY, U.S.A., *Department of Psychology, Smith College, Northampton, MA, U.S.A. and ‘University of Iowa and Iowa City Veterans Administration Medical Center, Iowa City, IA, U.S.A. (Received 4 September 1995)
Summary-Clinical evidence suggests that components of perfectionism may have special relevance to social phobia. This study examines this relationship by comparing 61 patients with social phobia and 39 community volunteers with no anxiety disorder on Frost, Marten, Lahart and Rosenblate’s (1990) Multidimensional Perfectionism Scale (MPS). Social phobia patients scored higher on subscales assessing concern over mistakes, doubts about actions, and perceived parental criticism. Community volunteers scored higher on the organization subscale. It was further hypothesized that, for the social phobia patients, perfectionism would be associated with greater symptom severity. Correlational analysis confirmed that the Concern over Mistakes and Doubts about Actions subscales of the MPS were consistently associated with greater social anxiety, trait anxiety, and general psychopathology. Implications for the treatment of social phobia patients are considered. Copyright 0 1996 Elsevier Science Ltd.
INTRODUCTION
Social phobia is a fear of situations in which an individual may be under the scrutiny of others and expects to act in an embarrassing or humiliating manner (DSM-IV; American Psychiatric Association, 1994). Individuals with social phobia often fear making unintelligent, unintelligible or socially inappropriate comments. They may fear that anxiety symptoms, such as blushing, sweating, or shaking, will be obvious to others, thereby ‘giving away’ that they are anxious. These and other fears about being the center of attention are often associated with doubt about one’s competence. Although we have learned much about social phobia in the last several years, we know little about personality dimensions which may influence the severity of patients’ distress or have implications for how they respond to various treatment strategies. This study examines one such dimension, perfectionism. Frost, Marten, Lahart and Rosenblate (1990) define perfectionism as the setting of excessively high personal standards plus a tendency to be overly self-critical upon failure to meet those standards. Frost et al. (1990) developed the Multidimensional Perfectionism Scale (MPS) to measure the key concepts of excessive personal standards and overconcern about making mistakes, and included other scales which have conceptual relevance to perfectionism including doubts about the quality of one’s actions, parental expectations, parental criticisms, and extreme organization. Perfectionism has been implicated as a factor in a variety of mental disorders including obsessivecompulsive disorder (e.g. McFall & Wollersheim, 1979), eating disorders (e.g., Garner, Garfinkel & Bemis, 1982) depression (Burns, 1980), and anxiety (e.g., Deffenbacher, Zwemer, Whisman, Hill & Sloan, 1986) and appears to have relevance to the clinical manifestations of social phobia. For example, our clinical experience suggests that individuals with social phobia often set virtually unattainable goals for themselves (excessive standards of performance) in feared social situations, thereby guaranteeing the occurrence of ‘mistakes’ in these situations. They also appear to make unreasonable comparisons when judging their own competence. Their public speaking efforts may be compared with the performance of television commentators or public officials while their social interactional behaviors are compared with those of the ‘life of the party’. When social phobic individuals do perform up to ‘par’, they are clearly not objective observers of their own behavior
*To whom all correspondence should be addressed at: Center for Stress and Anxiety Disorders, Pine West Plaza Building 4, Washington Avenue Extension, Albany, NY 12205, U.S.A.
404
Harlan R. Juster et af.
and will often devalue their otherwise solid performance or disqualify the situation as one that is atypical in some critical way (e.g. “I did well because the audience felt sorry for me” or “he was trying to be nice”) (Rapee & Lim, 1992). Furthermore, social phobic individuals engage in frequent avoidance of unfamiliar activities, possibly for fear of performing below standards. Heimberg and his colleagues (Heimberg & Becker, in press; Heimberg, Juster, Hope & Mattia, 1995) have developed a model of social phobia which may help explain the link between social phobia and perfectionism. Briefly, this mode1 proposes that a genetic susceptibility combines with early experience to sensitize individuals to view social encounters as threatening experiences. A set of (perfectionistic) beliefs develops which primes individuals to expect negative consequences in social situations, resulting in anxiety and often avoidance. Specifically, individuals with social phobia come to believe that (1) social situations are potentially dangerous (i.e. humiliating), (2) danger may be averted only by social performance above a very high standard (i.e. perfect performance), and (3) they are not capable of achieving the necessary standard to avoid danger. Achieving a certain standard may be hampered by over attention to mistakes in social situations. A vicious cycle may exist in which the expectation of mistakes produces excessive focus on evaluation of one’s performance which, in turn, causes a reduction in attention to the actual task, making mistakes more likely. Any mistake reduces the likelihood of attaining the desired standard, and therefore becomes an indicator of failure. Even when mistakes are within normal limits of magnitude or frequency for a particular situation, over attention to them may result in a biased view of the situation as a failure. Thus, excessive standards and concern over mistakes appear to be closely linked characteristics of individuals with social phobia. Although there is no direct evidence linking perfectionism and social phobia, several studies have examined the relationship among perfectionism, excessive standard setting, and social or performance anxiety in nonclinical populations. Each of these areas will be briefly considered. Frost and Henderson (1991) found that, for a sample of female college athletes, perfectionistic concern over mistakes was positively correlated with anxiety about performance in competition. Conversely, concern with mistakes and self-doubts were negatively correlated with self-confidence regarding athletic performance. Frost and Marten (1990) showed that perfectionistic college students, compared with nonperfectionistic students, experienced higher levels of negative affect while completing a writing task under conditions of high evaluative threat, a situation analogous to social encounters for social phobia patients. High perfectionistic students reported that doing well on the writing task was more important to them than it was to low perfectionistic students and believed they should have done better on the task as well. Test-taking is another area of academic performance which may activate evaluative anxiety. In fact, DSM-IV describes test anxiety as a variant of social phobia in which individuals fear indirect evaluation. Juster, Brown, Heimberg, Makris, Leung and Frost (1994) examined the association between Frost’s MPS and a measure of test anxiety in a sample of undergraduates. Predictably, the Concern over Mistakes and Doubts about Actions subscales were positively correlated with the cognitive and physiological responses associated with test anxiety. Mor, Day, Flett and Hewitt (1995) found that performance anxiety in professional musicians, actors, and dancers was positively correlated with perfectionistic personal and social standards. The largest association was found between socially prescribed perfectionism (i.e. standards based on others’ expectations) and anxiety. Clearly, perfectionism is associated with anxiety and other negative reactions in situations where individuals believe they are being evaluated. Negative reactions associated with concern over mistakes have received particular attention. This component of perfectionism is defined as negative cognitive reactions to mistakes, the tendency to equate mistakes with failure, and the expectation of negative responses from others regarding mistakes. Athletes high on concern over mistakes reported that they were more likely to focus on the negative aspects of their mistakes (Frost & Henderson, 1991). These include increased worry about others’ reactions, greater likelihood of feeling they let themselves down, more pressure to make up for the mistake, greater difficulty forgetting about it, and more recurrent images of the mistake. In another study, students scoring high on the Concern over Mistakes subscale of the MPS reported more negative affect and lower self-confidence, believed others judged them more harshly, and were more likely to conceal the details of their performance than low Concern over Mistakes
Social phobia and perfectionism
405
students after completing a task which elicited a high frequency of mistakes (Frost, Turcotte, Heimberg, Mattia, Holt & Hope 1995). Furthermore, high Concern over Mistakes students were more likely to believe that they ‘should’ have performed to a higher standard than their low-Concern over Mistakes counterparts. In contrast to the evidence regarding Concern over Mistakes, the Personal Standards subscale has not been associated with performance anxiety and the other negative reactions noted above, yet this construct has figured prominently in the definition of perfectionism and in our model of social phobia. In fact, Personal Standards has been associated with more positive strivings, as has been suggested by factor analysis (Frost, Heimberg, Holt, Mattia & Neubauer, 1993b). Alden, Bieling and Wallace (1994) noted that standards which exceed ability may be a motivating factor if the discrepancy between standards and ability is not too great. Thus, the effect of standards in evaluative situations may be complex. For example, Juster et al. (1994) found that Personal Standards interacted with the worry (cognitive) component of test anxiety to predict academic performance. Students scoring low on the worry component attained course grades directly related to their level of personal standards. High standards were associated with high academic performance and low standards were associated with poor performance. However, the academic performance of students scoring high on worry was unrelated to personal standards. Alden et al. (1994) showed that socially anxious undergraduates set personal standards in line with their abilities but lower than nonanxious students. Importantly, socially anxious students believed their abilities fell short of the standards set by others. Thus, personal standards may be lower in individuals with social phobia but still within their expected capabilities. However, they may believe their abilities fall short of the expectations that others hold for them. This is consistent with the clinical observation that social phobics can often perform tasks when alone without fear but are distressed by the same behavior when in the presence of others. The current study is the first investigation of the relationship of the dimensions of perfectionism and social phobia. We compared the MPS subscale responses of a group of social phobic subjects with those of a group of non-social phobic community volunteers. We predicted that social phobic subjects would score higher on this measure of perfectionism, most specifically the Concern over Mistakes subscale. For the social phobic patients only, the relationship of MPS subscale scores to measures relevant to social phobia and general psychopathology was examined. We predict that greater reliance on a perfectionistic style would be associated with greater severity of social phobia and that this relationship would not be better accounted for by patients’ score on a measure of depression.
METHOD Subjects Social phobic subjects. Sixty-one patients (38 males, 23 females) meeting DSM-III-R (American Psychiatric Association, 1987) criteria for social phobia participated in this study. Social phobic patients requesting treatment were identified by the Anxiety Disorders Interview Schedule-Revised (ADIS-R; DiNardo & Barlow, 1988). Subjects were rated on the O-8 Clinician’s Severity Rating (CSR), included in the ADIS-R, and only subjects who received a rating of four or greater (indicative of moderate impairment and need for treatment) were included in the study. The ADIS-R has been shown to have high rates of inter-rater agreement when social phobia is assigned as the principal diagnosis (k = 0.79; DiNardo, Moras, Barlow, Rapee & Brown, 1993). Patients were excluded from the study if they received a comorbid diagnosis of current major depression, bipolar disorder, psychotic disorder or active drug or alcohol dependence within the past three months. Participation in the study was part of the pretreatment assessment administered to all patients who enter treatment at the Social Phobia Program of the Center for Stress and Anxiety Disorders, University at Albany, State University of New York. Community volunteers. A nondisordered control sample of 2 1 women and 18 men from the Albany area were recruited by response to bulletin board advertisements requesting the assistance of persons who were neither anxious nor depressed. A detailed telephone screening interview was conducted with each respondent to rule out the presence of an anxiety disorder, affective disorder, or substance
406
Harlan R. Juster et al. Table I. Demographic characteristics for social phobia patients and community volunteers Social phobia patients (N=61) Mean age % Male % Married % Employed full-time % College graduate
36.6(9.6) 62 41 62 70
Community volunteers (N = 39) 34. I (10.0) 46 49 82 46
Figures in parentheses are standard deviations.
abuse or treatment for any of these conditions within the past two years. Control subjects were included in the order they passed screening, with the constraints that the sample be balanced for gender and equated for age with the social phobic sample. Demographic characteristics of the social phobic and community volunteer groups are summarized in Table 1. Measures The Multidimensional Perfectionism Scale (MPS; Frost et al., 1990) is a 35item self-report measure that contains six subscales related to the construct of perfectionism: (1) concern over mistakes (CM); (2) doubts about actions (DA); (3) personal standards (PS); (4) parental expectations (PE); (5) parental criticism (PC); and (6) organization (OR). The CM subscale is a measure of overly critical self-evaluation, a key concept in pathological perfectionism according to Frost et al. (1990). This scale includes items such as “People will probably think less of me if I make a mistake” and “If I fail at work/school, I am a failure as a person”. Another measure of critical self-evaluation, the DA subscale reflects a general dissatisfaction with or uncertainty about the quality of one’s effort or whether one has chosen the right course of action. Items in this subscale include “Even when I do something very carefully, I often feel that it is not quite right” and “It takes me a long time to do something right”. Of the MPS subscales, CM and DA have been most consistently related to measures of self-critical depression, compulsivity, procrastination and psychopathology (Frost et al., 1990). The PS subscale reflects a traditional view of perfectionism as the setting of excessively high standards and appears to have both positive and negative aspects (Frost et al., 1990). This is reflected in scale items such as “It is important for me to be thoroughly competent in everything I do” and “I expect higher performance in my daily tasks than most people”. Expectations and critical evaluations of parents towards their children have a role in perfectionism and may be etiologically associated with personal expectations and self-evaluations (Frost, Lahart & Rosenblate, 1991). An example of an item from each of these scales is “My parents wanted me to be the best at everything” (PE) and “My parents never tried to understand my mistakes” (PC). The OR subscale reflects an individual’s tendency to emphasize orderliness and precision in daily tasks. OR scale items include “Organization is very important to me” and “I am a neat person”. Reliability and validity of the MPS. Internal consistency has ranged from good to excellent for each of the subscales (US= 0.77-0.93) and the total perfectionism score (c( = 0.90; Frost et al., 1990). The total perfectionism score (equal to the sum of all subscales except OR) was significantly correlated with other measures of perfectionism while CM and DA have been most consistently correlated with measures of general psychopathology (Frost et al., 1990). CM was directly related to procrastination while PS and OR were correlated with lower frequencies of procrastination, supporting speculation about the positive aspects of these two scales. As noted previously, MPS scores have predicted subjects’ reactions to evaluative threat (Frost & Marten, 1990) and to making mistakes (Frost et al., 1995). Other assessments. Instruments measuring various aspects of social and general anxiety and general psychopathology were administered to the social phobic patients. The Social Interaction Anxiety Scale (SIAS; Mattick & Clarke, 1989) assesses anxiety in social interaction situations while the Social Phobia Scale (SPS; Mattick & Clarke, 1989) focuses primarily on anxiety while being observed. The Social Phobia subscale of the Fear Questionnaire (FQSO; Marks & Mathews, 1979)
Social phobia and perfectionism
407
provides a measure of avoidance of typical social situations and the Fear of Negative Evaluation Scale (FNE; Watson & Friend, 1969) addresses patient’s concerns about the disapproval of others. Also administered were the Trait form of the State-Trait Anxiety Inventory (STAI-T; Spielberger, Gorsuch, Lushene, Vagg & Jacobs, 1983) the Global Symptom Index (GSI) of the Symptom Checklist-90-Revised (SCL-90-R; Derogatis, 1977), and the Beck Depression Inventory (BDI; Beck, Ward, Mendelson, Mock & Erbaugh, 1961). The interviewer-administered CSR and the Fear (LSAS-F) and Avoidance (LSAS-A) Subscales of the Liebowitz (1987) Social Anxiety Scale were included as interviewer-rated measures of social anxiety.
RESULTS Demographic characteristics
Social phobic patients were compared with community volunteers on several demographic variables. There were no significant differences between the two groups on gender, age, or marital status. They differed on educational level, such that a greater percentage of social phobic patients had graduated from college (Yates corrected x2 (1, N = 100) = 4.94, P = 0.026). Employment level was examined using a number of different groupings. When the number of individuals employed fulltime was compared with all other categories combined (part-time, student, homemaker, and unemployed), there was a trend for a greater percentage of community volunteers to be employed on a full-time basis (Yates corrected x2 (1, N = 100) = 3.53, P = 0.06). There were no differences when full- and part-time employment were combined and compared with the remaining groups. A series of analyses of variance revealed that neither educational level nor employment status was related to any of the MPS subscales, so these variables are not considered further in the analyses of our findings. Furthermore, males and females with social phobia did not differ on any MPS subscales, a finding consistent with Frost, Heimberg, Holt and Mattia’s (1993a) results comparing a large sample of undergraduates. Male and female social phobics also showed no differences on any measures of social anxiety or general psychopathology. All subsequent analyses combine these two groups. MPS subscales
Differences between the patients and community volunteers on MPS subscale scores were evaluated by multivariate analysis of variance. With the six subscales scores as the dependent variables, the multivariate test was significant (Wilks’ i = 0.793; F(4,93) = 4.03, P < 0.01). Follow-up univariate analyses of variance indicated that social phobics and community controls differed on CM, DA, PC, and OR but not on PS or PE. Means, standard deviations, and results of univariate F-tests are provided in Table 2. For CM, DA, and PC, social phobia patients scored higher than community controls. The community controls scored higher than the social phobics on OR.
Table 2. Means, standard deviations, univariate F-ratios, and probability values for Multidimensional Perfectionism Scale scores and differences between groups
MPS subscale CM PS DA PC PE OR
Social phobia patients (N = 61)
Community volunteers (N = 39)
F
P
25.6(7.5) 23.1(5.8) 10.9(3.7) 9.3 (3.9) 12.7(6.0) 20.1(5.4)
20.3(6.7) 23.4(4.9) 9.0(3.5) 7.8(3.1) 12.5(3.9) 23.1(5.0)
12.83 0.07 7.21 4.15 0.06 7.67
0.001 o.“do9 0.044 o.noso7
MPS = Multidimensional Perfectionism Scale; CM = Concern Over Mistakes; PS = Personal Standards; DA = Doubts About Actions; PC = Parental Criticism; PE = Parental Expectations; OR = Organization. Figures in parentheses are standard deviations.
408
Harlan R. Juster ef al.
Table 3. Zero-order and semi-partial correlations between MPS subscales and measures of social and general anxiety for social phobia patients
CM
PS
MPS subscales DA
PE
PC
OR
0.25 0. I5 0.07 0.05
-0.05 -0.08 0.22 0.00
Social an.riet.v Se@eport
Fear questionnaire-Social Phobia Social interaction anxiety scale Social phobia scale Fear of negative evaluation scale
0.46’ 0.36 0.26 0.50;
(0.38t)
(0.39t)
-0.02 -0.03 0.08 0.00
0.44* 0.52’ 0.36 0.43’
(0.30) (0.33t) (0.27)
0.12 0.16 0.04 0.16
-0.05 0.08 0.03
0.39’ 0.531 0.50*
(0.28) (0.37t) (0.35t)
0.07 0.23 0.20
0.06 0.26 0.16
0.02 -0.03 0.05
0.09 0.1 I 0.19
0.37 0.61’ 0.47*
(0.37t) (0.21)
0.07 0.07 0.09
0.16 0.24 0.21
-0.06 0.02 0.07
Intemiewer-administered
ADIS-R clinician’s severity rating Liebowitz social anxiety scale-fear Liebowitz social anxiety scale-avoidance
0.21 0.44* 0.36
General psvchopathology Beck depression inventory State-trait anxiety inventory-trait SCL-90-R global symptom index
0.24 0.48* 0.40’
(0.33t)
(0.3lt) (0.23t)
MPS = Multidimensional Perfectionism Scale; CM = Concern Over Mistakes; PS = Personal Standards; DA = Doubts About Actions; PC = Parental Criticism; PE = Parental Expectations; OR = Organization; SCL-90-R = Symptom Checklist 90-Revised; ADISR = Anxiety Disorders Interview Schedule-Revised Values in parentheses are semi-partial correlations controlling for Beck Depression Inventory score. For zero-order correlations, * Indicates P < 0.002 for self-report social anxiety measures and P < 0.003 for all other measures. For semi-partial correlations. t indicates P < 0.01 for CM and P -c 0.0063 for DA.
MPS, social anxiety, andpsychopathology
The relationship between perfectionism, social anxiety, and general psychopathology was examined for the social phobic patients. Correlations between MPS subscales and other measures were calculated. Measures were classified as self-reports of social anxiety, interviewer assessments of social anxiety, or self-reports of general psychopathology. Given the large number of zero-order correlations, Bonferroni corrections were applied to each group of measures. Thus, for the four self-report measures of social anxiety, probabilities of 0.002 or less were considered significant. For interviewer-administered measures of social anxiety and self-report measures of general psychopathology (three measures each), probability values of 0.003 or less were considered significant. Correlations are provided in Table 3. Several significant correlations were found between CM and DA and measures of social anxiety and psychopathology. CM was significantly correlated with FQSO, FNE, STAI-T, GSI, and the LSAS-F. DA was significantly correlated with all measures except the SPS and the BDI. No significant zero-order correlations were found between PS, PE, PC, OR subscales and any other measure. Significant zero-order correlations were followed by examination of semi-partial correlations controlling for the effect of depression (BDI). Semi-partial correlations were examined since depression is correlated with most measures of anxiety and our interest was in the unique relationship between dimensions of perfectionism and social anxiety. Once again, Bonferroni corrections were applied. For CM, semi-partial correlations with a probability of 0.01 or less were considered significant, whereas for DA, probabilities of less than 0.0063 were considered significant. All correlations involving CM and four of eight correlations with DA remained significant after controlling for depression. These findings generally confirm the pattern of relations between CM, DA, and measures of social anxiety and general psychopathology, although the findings for DA are somewhat less robust than suggested by the zero-order correlations.* *The helpful comments of an anonymous reviewer to an earlier version of this manuscript suggest an alternati-/e strategy. First, a single composite factor score was developed composed of the self-report and interviewer-generated social anxiety measures. The alpha reliability for this composite of seven measures was 0.88. Hierarchical multiple regression was performed on the composite score which was regressed on BDI and SCLGSI entered as the first block of the regression equation (TRAIT was not included in this block as it would have reduced the variance available for prediction to practically nil). The second block (the one of interest in this analysis) was composed of the CM, DO, and PS subscales of the MPS. The second block produced a significant increment in variance accounted for over the first block (Fchange = 3.95, P < 0.02) indicating that the linear combination of CM, DO, and PS accounted for a significant proportion of variance in the composite social anxiety measure beyond that predicted by a measure of depression and general psychopathology. Examination of the semi-partial correlations for the MPS subscales (0.16, 0.17, and - 0.17 for CM, DO, and PS, respectively) indicate virtually equivalent contributions of these predictors although all three failed to reach significance independently (P = 0.08) despite their significant joint contribution.
Social phobia and perfectionism
409
DISCUSSION Patients with social phobia were more likely to achieve extreme scores on a measure of perfectionism than a sample of community volunteers. Social phobics were more concerned about making mistakes and more likely to doubt their own actions; they reported more parental admonishment as a result of not living up to parent’s standards and less concern about orderliness and organization than community volunteers without an anxiety disorder. These empirical findings support our clinical observations that social phobic patients expect to make critical errors and doubt their abilities in social interaction and public performance. They tend to focus more attention on their mistakes, overvalue the importance of mistakes, and blame themselves for poor outcomes. Social phobic patients scored higher on the Concern over Mistakes subscale than the community volunteers. Concern over Mistakes is associated with increased negative affect, the interpretation of mistakes as a sign of specific and global failure, concomitant reduction in motivation to persist at an activity, and the assumption that others also see them as a failure. Thus, the stage is set for the cycle of fear and avoidance of social situations which characterizes social phobia. Social phobic individuals also appear to devalue their performance and experience much self-doubt (as indicated by their higher scores on the Doubts about Actions subscale) when they have chosen a particular course of action. Self-doubt may focus on one’s ability to perform required social behaviors, in giving the ‘correct’ response or asking the ‘proper’ questions in a social interaction, or in meeting personal goals. Consistent with the model described earlier (Heimberg et al., 1995). social anxiety occurs when individuals believe they will be unable to create a desired impression in a social situation (Leary & Atherton, 1986; Schlenker 8c Leary, 1982). For the social phobic patients, greater concern over mistakes and doubts about actions were indicative of more severe social anxiety, trait anxiety, and general psychopathology. The associations remained even after the effect of depression was statistically controlled. Notably, Doubts about Actions and Concern over Mistakes were also correlated with interviewer-rated measures. The finding of an association between components of perfectionism and social anxiety z:cross different modes of assessment lends further credence to this relationship. Although social phobics reported more parental criticism than community volunteers, they reported similar levels of parental expectations. Conceptually, parental criticism appears to be more closely associated with other negative aspects of perfectionism, especially concern over mistakes. Expectations, however, appear to be more highly associated with personal standards, a measure of self-expectations. This pattern of relationships is supported by interscale correlations from a previous study (Frost et al., 1990). Parental criticism may have etiological significance for social phobia. Several studies retrospectively examined the parental-rearing styles in the families of social phobic patients (Arrindell, Emmelkamp, Monsma & Brilman, 1983; Parker, 1979). Bruch, Heimberg, Berger and Collins (1989) showed that social phobic patients perceived their parents as placing greater emphasis on the opinions of others regarding appropriate behavior than agoraphobic patients. Bruch and Heimberg (1994) replicated this finding and further demonstrated that parents of social phobics relied more on the use of shame as a disciplinary technique than other parents. They further argue that this pattern of repeatedly sensitizing children to the scrutiny of others is a key element in etiological mechanisms of social anxiety (Buss, 1980). The two groups did not differ on their level of personal standards, a finding consistent with previous work by Frost and his colleagues. This dimension of perfectionism has both positive and negative elements and may be common both among individuals who set and successfully strive for high goals and those who set goals beyond their capacity (Frost et ul., 1993b). As noted earlier, the distinguishing feature may be the magnitude of the discrepancy between standards and perceived ability (Alden et al., 1994). A limitation of this study is that the only comparison group consisted of nondisordered community volunteers. Thus, although the effects of depression were statistically controlled, it is not clear how the current pattern of findings might have differed had an alternative group of patients, for example with panic disorder, been included for comparison. Future research might also examine whether differences in patients’ perfectionistic styles are predictive of treatment outcome and whether treatment can alter those styles.
410
Harlan R. Juster et al.
Acknowledgements --Preparation
of this article was supported in part by grant 44119 to Richard G. Heimberg from the National Institute of Mental Health. Portions of this paper were presented at the annual meeting of the Association for Advancement of Behavior Therapy, Boston, MA, November 1992. Jill Mattia is now at the Deiartment of Psychiatry, Brown Medical School at Rhode Island Hospital, Providence, Rhode Island.
REFERENCES Alden, L. E., Bieling, P. J. & Wallace, S. T. (1994). Perfectionism in an interpersonal context: A self-regulation analysis of dysphoria and social anxiety. Cognitive Therapy and Research, 18, 291-316. American Psychiatric Association (1987).Diagnostic and Statistical Manual of Mental Disorders (3rd rev. edn). Washington, DC: Author. American Psychiatric Association (1994).Diagnostic and Statistical Manual of Mental Disorders (4th edn). Washington, DC: Author. Arrindell, W. A., Emmelkamp, P. M. G., Monsma, A. & Brilman, E. (1983). The role of perceived parental rearing practices in the etiology of phobic disorders: A controlled study. British Journal of Psychiatry, 143, 183-187. Beck, A. T., Ward, C. H., Mendelson, M., Mock, J. & Erbaugh, J. (1961). An inventory for measuring depression. Archives of General Psychiatry, 4, 561-511.
Bruch, M. A. & Heimberg, R. G. (1994). Differences in perceptions of parental and personal characteristics between generalized and nongeneralized social phobics. Journal of Anxiety Disorders, 8, 155-168. Bruch, M. A., Heimberg, R. G., Berger, P. & Collins, T. M. (1989). Social phobia and perceptions of early parental and personal characteristics. Anxiety Research, 2, 5765. Burns, D. (1980). The perfectionist’s script for self-defeat. Psychology Today, November, 34-51. Buss, A. H. (1980). Self-consciousness and social anxiety. San Francisco: Freeman. Deffenbacher, J. L., Zwemer, W. A., Whisman, M. A., Hill, R. A. & Sloan, R. D. (1986). Irrational beliefs and anxiety. Cognititre Therapy and Research, IO, 281-292.
Derogatis, L. R. (1977). SCL-90 administration, scoring and procedures manual for the revised version. Baltimore: Johns Hopkins University School of Medicine. DiNardo, P. A. & Barlow, D. H. (1988). The Anxiety Disorders Interuiew Schedule-Revised(ADIS-R). Albany, NY: Graywind. DiNardo, P. A., Moras, K., Barlow, D. H., Rapee, R. M. & Brown, A. A. (1993). Reliability of DSM-III-R anxiety disorder categories: Using the Anxiety Disorders Interview Schedule-Revised. Archives of General Psychiatry, 50, 251-256. Frost, R. 0.. Heimberg, R. G., Holt, C. S. & Mattia, J. I. (1993a). Perfectionism andgender. Unpublished manuscript. Frost, R. O., Heimberg, R. G., Holt, C. S., Mattia, J. I. & Neubauer, A. L. (1993b). A comparison of two measures of perfectionism. Personality and Individual Differences, 14, 119-126. Frost. R. 0. & Henderson, K. J. (1991). Perfectionism and reactions to athletic competition. Journal of Sport and Exercise Psycholog.y, 13, 323-335.
Frost, R. O., Lahart, C. M. & Rosenblate, R. (1991). The development of perfectionism. Cognitive Therapy and Research, IS, 469490.
Frost, R. 0. & Marten, P. A. (1990). Perfectionism and evaluative threat. Cognitive Therapy and Research, 14, 5599572. Frost, R. O., Marten, P. A., Lahart, C. & Rosenblate, R. (1990). The dimensions of perfectionism. Cognitiue Therapy and Research, 14,449-468.
Frost, R. O., Turcotte, T. A., Heimberg, R. G., Mattia, J. I., Holt, C. S. & Hope, D. A. (1995). Reactions to mistakes among subjects high and low in perfectionistic concern over mistakes. Cognitive Therapy and Research, 19, 195-205. Garner, D., Garfinkel, P. & Bemis, K. (1982). A multidimensional psychotherapy for anorexia nervosa. International Journal of Eating Disorders, I, 3-64.
Heimberg, R. G. & Becker, R. E. (in press). Treatment of socialfears andphobias. New York: Guilford. Heimberg, R. G., Juster, H. R., Hope, D. A. & Mattia, J. I. (1995). Cognitive-behavioral group treatment for social phobia: Description, case presentation, and empirical support. In Stein, M. (Ed.) Socialphobia: Clinical and research perspectives (pp. 293-321). American Psychiatric Press. Juster, H. R., Brown, E. J., Heimberg, R. G., Makris, G. S., Leung, A. W. & Frost, R. 0. (1994). Effect of test anxiety and perfectionism on academic performance. Poster presented at the annual meeting of the Association for Advancement of Behavior Therapy, San Diego, CA. Leary. M. R. & Atherton, S. C. (1986). Self-efficacy, social anxiety, and inhibition in interpersonal encounters. Journal of Social and Clinical Psychology, 4, 256267.
Liebowitz, M. R. (1987). Social phobia. Modern Problems of Pharmacopsychiatry, 22, 141-173. Marks, I. M. & Mathews, A. M. (1979). Brief standard self-rating for phobic patients. Behaviour Research and Therapy, 17, 263-267.
Mattick, R. P. & Clarke, J. C. (1989). Development and validation of measures of social phobia scrutiny fear and social interaction anxiety. Unpublished manuscript. McFall, M. E. & Wollersheim, J. P. (1979). Obsessive-compulsive neurosis: A cognitive-behavioral formulation and approach to treatment. Cognitive Therapy and Research, 3, 333-348. Mor, S., Day, H. I., Flett, G. L. & Hewitt, P. L. (1995). Perfectionism, control and components of performance anxiety in professional artists. Cognitive Therapy and Research, 19, 207-225. Parker, G. (1979). Reported parental characteristics of agoraphobics and social phobic% British Journal of Psychiatry, 135, 555-560.
Rapee, R. M. & Lim, L. (1992). Discrepancy between self- and observer ratings of performance in social phobics. Journal of Abnormal Psychology, 101, 727-731.
Schlenker, B. S. & Leary, M. R. (1982). Social anxiety and self-presentation: A conceptualization and model. Psychological Bulletin, 92, 64-669.
Spielberger, C., Gorsuch, R., Lushene, R., Vagg, P. & Jacobs, G. (1983). Manualfor the State-Trait Anxiety Inventory. Palo Alto, CA: Consulting Psychologists Press. Watson, D. & Friend, R. (1969). Measurement of social-evaluative anxiety. Journal of Consulting and Clinical Psychology, 33,448-457.