Construct validation of a state version of the Social Physique Anxiety Scale among young women

Body Image 8 (2011) 52–57

Contents lists available at ScienceDirect

Body Image journal homepage: www.elsevier.com/locate/bodyimage

Construct validation of a state version of the Social Physique Anxiety Scale among young women Kathleen A. Martin Ginis ∗ , Elisa Murru, Catherine Conlin, Heather A. Strong McMaster University, Hamilton, ON L8S 4K1, Canada

a r t i c l e

i n f o

Article history: Received 15 July 2010 Received in revised form 1 October 2010 Accepted 1 October 2010 Keywords: Exercise Measurement Body image Self-presentation Social physique anxiety

a b s t r a c t This study examined the validity of a state version of the Social Physique Anxiety Scale (Hart, Leary & Rejeski, 1989) by conducting tests of concurrent and discriminative validation. Participants were four separate samples of young women (N = 221) who exercised ≤2 days/week and who participated in various experiments examining body image and self-presentation. Participants’ scores on the state SPAS (S-SPAS) were significantly correlated, in expected directions, with scores on both trait and state measures of body image and self-presentation, and with body mass index (BMI). In addition, S-SPAS scores discriminated between women who exercised in a mixed-sex versus a same-sex environment, but trait SPAS scores did not. Together, these results provide evidence of construct validity of a state version of the SPAS and demonstrate that social physique anxiety can be conceptualized and measured as a situational variable. The S-SPAS, rather than the trait SPAS, should be employed in experiments designed to detect differences in state social physique anxiety. © 2010 Elsevier Ltd. All rights reserved.

Introduction Social physique anxiety refers to the anxiety people experience when they are worried about others’ evaluations of their bodies (Hart, Leary, & Rejeski, 1989). Twenty years ago, Hart and colleagues developed the Social Physique Anxiety Scale (SPAS) to measure people’s tendencies to experience social physique anxiety. Since then, the SPAS has been used in dozens of studies, typically as a predictor of exercise and other health-related behaviours or as an outcome of interventions designed to improve body image (for reviews, see Hausenblas, Brewer, & Van Raalte, 2004; Martin Ginis & Leary, 2004; Martin Ginis, Lindwall, & Prapavessis, 2007). As noted by other investigators, social physique anxiety was originally conceptualized as a relatively stable personality trait (Marquez & McAuley, 2001; McAuley, Bane, & Mihalko, 1995). However, it has become increasingly recognized that social physique anxiety can fluctuate across situations. For instance, scenario studies have shown that women anticipate experiencing greater social physique anxiety in mixed-sex than all-female exercise classes (Kruisselbrink, Dodge, Swanburg, & MacLeod, 2004), and in exercise classes led by a male instructor rather than a female instructor (Amirthavasar & Bray, 2007). Furthermore, women who perceive their fitness instructor to be more attractive than them-

∗ Corresponding author at: McMaster University, Department of Kinesiology, Hamilton, ON L8S 4K1, Canada. Tel.: +1 905 525 9140x23574; fax: +1 905 523 6011. E-mail address: [email protected] (K.A. Martin Ginis). 1740-1445/$ – see front matter © 2010 Elsevier Ltd. All rights reserved. doi:10.1016/j.bodyim.2010.10.001

selves experience greater situational social physique anxiety than those who perceive their instructor to be just as attractive or less attractive than they are (Martin Ginis, Prapavessis, & Haase, 2008). Taken together, these studies support the notion that in addition to having dispositional qualities, social physique anxiety can also have situational manifestations and fluctuations. Conceptualizing social physique anxiety as both a dispositional and state variable is consistent with conceptualizations of the parent concepts of self-presentation and body image. Self-presentation (also known as “impression management”) refers to the processes people use to monitor and control how they are perceived by others (Schlenker, 1980). According to the Two-Component Model of Impression Management and its supporting research (Leary & Kowalski, 1990), self-presentational concerns can vary across people and situations. Given that social physique anxiety is an affective manifestation of self-presentational concerns about one’s body, it makes sense that social physique anxiety would also vary both between and within individuals. Body image is a multidimensional concept that captures thoughts, feelings, behaviours, and perceptions regarding one’s body. Although body image was originally conceptualized as a trait, accumulating data indicate that body experiences can vary across situations (Cash, Fleming, Alindogan, Steadman, & Whitehead, 2002; Haimovitz, Lansky, & Oreilly, 1993; Wasilenko, Kulik, & Wanic, 2007). As an affective expression of body image (Bane & McAuley, 1998), it is understandable that social physique anxiety would also vary across situations. In order to accurately detect and measure momentary changes in body experiences, state measures are needed (Cash et al., 2002).

K.A. Martin Ginis et al. / Body Image 8 (2011) 52–57

To date, investigators have assessed state manifestations of social physique anxiety by modifying the original trait SPAS instructions and items to make them relevant to their individual studies (e.g., Amirthavasar & Bray, 2007; Kruisselbrink et al., 2004; Martin Ginis et al., 2008). Although these modified scales have successfully detected differences in state social physique anxiety across various contextual manipulations, construct validation of these measures has not yet been undertaken. Furthermore, without a standardized set of state SPAS items, different investigators may make different modifications to the scale items, which can ultimately lead to difficulties comparing findings across studies. Given these concerns, the primary purpose of the present study was to investigate a state version of the SPAS and examine its construct validity. Construct validation is an on-going process of making new predictions about a construct and then testing them using various approaches (Streiner & Norman, 2008). In the present study, we undertook tests of convergent and discriminative (also known as “extreme groups”) validation. With regard to convergent validation, it was hypothesized that scores on the state SPAS (S-SPAS) would be significantly correlated with measures of body image and self-presentation, and with body mass index (BMI). Discriminative validation involves administering the scale to two groups who would be expected to have significantly different scale scores (Streiner & Norman, 2008). We compared S-SPAS scores among women exposed to a same-sex versus a mixed-sex exercise environment. Based on previous scenario research (Kruisselbrink et al., 2004), it was hypothesized that women who exercised in a mixedsex environment would have higher state social physique anxiety than those exercising in a same-sex environment. A secondary study purpose was to compare the responsiveness of the S-SPAS versus the trait SPAS in a situation expected to elicit social physique anxiety. Responsiveness is another aspect of validation, and refers to the ability of an instrument to measure meaningful change in an attribute (Streiner & Norman, 2008). We compared the responsiveness of the trait and S-SPAS to determine whether a state version of the SPAS is necessary to detect situational fluctuations in social physique anxiety. It was hypothesized that differences in state, but not trait SPAS scores would emerge among women who exercised in a mixed-sex versus same-sex environment. Method Participants The data presented in this article were collected in four separate studies involving samples of young women who participated in ≤2 bouts/week of moderate-heavy intensity leisure time physical activity. Sample 1 (n = 47) consisted of data not previously published. Data from the other three samples constitute secondary analyses. Samples 2 (n = 50; Martin Ginis, Strong, Arent, & Bray, under review) and 3 (n = 44; Strong, 2010) consisted of women who participated in studies examining stress responses in different social evaluative situations. Sample 4 (n = 80) consisted of women who participated in a study of the effects of exercise videos on body image (Martin Ginis et al., 2008). Characteristics of the study samples are presented in Table 1. Measures Trait Social Physique Anxiety Scale (SPAS; Hart et al., 1989). Originally, the SPAS was developed as a 12-item measure. However, concerns about the factor structure and some negatively worded items prompted various investigators to recommend the deletion of scale items (for a review of these issues, see Martin Ginis et al.,

53

2007). Consequently, a team of investigators that included two of the original SPAS authors (Leary and Rejeski) recommended a 9-item version of the SPAS (Martin, Rejeski, Leary, McAuley, & Bane, 1997). This team showed that in samples of younger and adult women, after deleting three of the original SPAS items, the scale’s unidimensional factor structure was supported, the items had good internal consistency (˛ = .89), and total SPAS scores were correlated in meaningful ways with related concepts. As such, the 9-item version of the SPAS was administered in the present study, as a measure of trait social physique anxiety. Participants indicated the degree to which each statement was characteristic or true of themselves using a 5-point Likert scale (1 = not at all; 5 = extremely). SPAS items are shown in the first column of Appendix A. Items 2 and 8 were reverse scored and a total SPAS score was then calculated. Higher scores indicate greater tendencies to experience social physique anxiety. State Social Physique Anxiety Scale (S-SPAS). The S-SPAS is a modified version of the 9-item (Martin et al., 1997) Social Physique Anxiety Scale (SPAS; Hart et al., 1989) and is designed to capture situational social physique anxiety. Given the history of the SPAS, described above, the 9-item version of the SPAS was used as the basis for creating the S-SPAS rather than the original 12-item version. The SPAS items were originally modified by Kruisselbrink and Martin Ginis in 2001 for the purpose of measuring state social physique anxiety in exercise settings. Pilot testing of the revised items was undertaken in Martin Ginis’s lab (Burke, 2002). The items were subsequently refined and used in Kruisselbrink’s (Kruisselbrink et al., 2004) study to examine men’s and women’s responses to hypothetical scenarios describing mixed- and samesex exercise environments. As shown in Appendix A, most of the items required only minor rewording to make them situationally relevant. However the 9th item, which was originally phrased to assess physique anxiety about wearing a bathing suit, was rewritten to assess physique anxiety experienced while wearing workout clothes. This item was phrased somewhat differently across samples, depending on what participants were required to wear during the experiment (i.e., “workout clothes,” “usual workout clothes,” or “shorts and a t-shirt”), and in order to ensure that the item was meaningful within the experimental context. Likewise, the instruction set for the scale varied slightly across the four studies to ensure that the instructions were appropriate and meaningful in each experimental context. In two studies (Samples 1 and 2), the measure was administered immediately after women exited an exercise environment, so the questions were written in the past tense and preceded by the following instructions: “Indicate the number that best represents the extent to which you experienced the feelings described by each item.” In one study (Sample 3), the scale was administered immediately after body fat measurements were conducted on study participants, and while participants were still in the assessment environment. Thus, the items were written in the present tense and the instructions stated: “Indicate the number that best represents the extent to which you are experiencing the feelings described by each item.” In the fourth study, participants exercised to a workout video in an isolated environment, so they were instructed to respond to the S-SPAS by imagining themselves in a real-life aerobics class (a methodology analogous to the scenario studies that have been used previously to study state social physique anxiety; Amirthavasar & Bray, 2007; Kruisselbrink et al., 2004). The items were written in the conditional verb tense (e.g., “I would feel uptight about my physique/figure”) and items that referred to “other people” were replaced with “the instructor” to adapt the S-SPAS to the imagined, instructor-led aerobics setting. Responses were made on 5-point Likert-type scales (1 = not at all; 5 = a great deal). For grammatical reasons, the anchor for “5” was changed to “a great deal” in the S-SPAS from “extremely” in

54

K.A. Martin Ginis et al. / Body Image 8 (2011) 52–57

Table 1 Characteristics of the studies’ samples.

n Age Ethnicity (%) Caucasian Asian Black Other BMI Bouts of activity per week Mild Moderate Vigorous Trait SPA

Sample 1

Sample 2

Sample 3

Sample 4

47 21.70 ± 3.45

50 20.78 ± 2.48

44 21.61 ± 2.92

80 26.49 ± 7.41

53 21 2 23 22.64 ± 4.70

80 12 4 4 22.35±5.60

80 2 5 14 22.96 ± 3.89

44 49 0 7 22.13 ± 3.26

2.21 ± 2.64 0.81 ± 0.90 0.70 ± 1.03 29.34 ± 8.21

2.26 ± 2.56 0.62 ± 0.70 0.76 ± 0.80 26.56 ± 7.39

1.11 ± 1.33 0.68 ± 0.56 0.57 ± 0.59 33.05 ± 5.27

1.90 ± 2.51 0.74 ± 1.11 0.94 ± 1.32 26.01 ± 6.71

the original SPAS. Given that the Body Image States Scale (Cash et al., 2002) employs “extremely” for its extreme anchor in some items and “a great deal” for other items, this modification seemed reasonable. Items 2 and 8 were subsequently reverse scored and a total scale score was calculated. Higher scores indicate higher state social physique anxiety. Self-Presentational Efficacy Scale (Gammage, Hall, & Martin Ginis, 2004). This 5-item scale assesses people’s confidence in their abilities to convey the impression that they are fit and competent exercisers. Participants were instructed to think about exercising in a public setting and to rate their confidence that other people in the setting would think that they had good physical coordination, good stamina, that their bodies looked fit and toned, that they exercise regularly, and that they were in good shape. Responses were made on a scale ranging from 0% (not at all confident) to 100% (completely confident). The scale was scored by averaging across items. Higher scores indicate greater self-presentational efficacy in exercise environments. Physical Appearance State-Trait Anxiety Scale (Reed, Thompson, Brannick, & Sacco, 1991). The state version of PASTAS was administered in this study, and consisted of two 8-item subscales that separately measure weight-related and non-weight related aspects of physical appearance anxiety. Participants used a 5-point scale (0 = not at all; 4 = exceptionally so) to rate the extent to which they felt anxious, tense, or nervous about weight-related (e.g., “my thighs”) and non-weight-related aspects of their bodies (e.g., “my ears”). Higher total scores indicate higher state appearance anxiety. Body Image Avoidance Questionnaire (Rosen, Srebnik, Saltzberg, & Wendt, 1991). This scale is a 19-item self-report instrument that measures the tendency to avoid physical appearance-salient situations. Participants used a 6-point scale (0 = never; 5 = always) to indicate the frequency with which they engaged in behaviours associated with body-image disturbance (e.g., “I wear baggy clothes” and “I do not go out socially if it involves eating”). Items were summed to obtain a total score, with higher scores indicating greater levels of body-image disturbance. Body Image States Scale (BISS; Cash et al., 2002). The BISS is a 6-item scale that measures affective and evaluative aspects of state body image. Participants responded to six items that assessed how they felt, at that moment, regarding their physical appearance, body size and shape, weight, physical attractiveness, their looks, and how they look relative to the average person. Each item had 9 response options. Responses were summed such that higher scores indicate a more positive state body image. Multidimensional Body Self-Relations Questionnaire (MBSRQ; Brown, Cash, & Mikulka, 1990; Cash, 2000). The MBSRQ consists of 10 subscales that measure dispositional aspects of body image. Four MBSRQ subscales were administered in this study—Self-Classified Weight, Body Areas Satisfaction, Appearance Evaluation, and Fitness Evaluation.

The 2-item Self-Classified Weight subscale is a measure of perceptions of one’s own weight. Participants chose one of five statements (1 = very underweight; 5 = very overweight) to complete two statements regarding their weight (e.g., “I think I am. . .” and “From looking at me, most other people would think I am. . .”). The two items were averaged, with higher scores indicating perceptions of being more overweight. The 9-item Body Areas Satisfaction Scale is a measure of satisfaction with discrete areas of one’s body and physical appearance. Participants used a 5-point Likert-type scale (1 = very dissatisfied; 5 = very satisfied) to rate how satisfied they were with different aspects of their body (e.g., “Face [facial features, complexion]” and “Muscle tone”). An average score was calculated across the ratings. The Appearance Evaluation subscale is a 7-item measure of feelings of physical attractiveness (e.g., “My body is sexually appealing” and “I like my looks just the way they are”). Participants used a 5-point scale (1 = definitely disagree to 5 = definitely agree) to respond to each item. Two items were reverse scored, and all items were averaged to calculate a total subscale score. The Fitness Evaluation subscale is a 3-item measure of perceptions of being physically fit (e.g., “I easily learn physical skills” and “I do poorly in physical sports or games”). Participants used a 5-point scale (1 = definitely disagree to 5 = definitely agree) to respond to each item. After reverse scoring one item, the three items were averaged to obtain a total score. Body Mass Index (BMI). For Samples 1, 2, and 4, BMI was calculated (kg/m2 ) from participants’ self-reported height and weight. For Sample 3, BMI was calculated from the experimenter’s direct measure of each participant’s height and weight. Procedure Convergent validity. Participants completed the S-SPAS, SPAS, and at least one other measure as part of the experimental protocol. As shown in Table 2, different samples completed different measures. In each study, the state measures were administered at the same time as the S-SPAS, and immediately after exposure to an experimental manipulation. Questionnaires were administered by one of the investigators and were completed in privacy. All participants provided informed consent and the study protocols were approved by the researchers’ university research ethics boards. Discriminative validity and responsiveness. A between-groups, pretest/posttest experimental design was used to examine the discriminative validity of the S-SPAS and the responsiveness of the trait SPAS versus the S-SPAS. The study protocol was approved by the McMaster University research ethics board. Participants were the 47 women recruited for Sample 1 (see Table 1) for a study advertised as an examination of women’s thoughts and feelings toward a beginner’s strength-training class. Participation was limited to women who had never followed a regular strength-training program (defined as 2–3 bouts of strength-training per week for at least

K.A. Martin Ginis et al. / Body Image 8 (2011) 52–57 Table 2 Bivariate correlations between the State Social Physique Anxiety Scale (S-SPAS) and other measures of self-presentation, body image, and body mass index. Scale

n

r

Sample

Trait Social Physique Anxiety

47 50 44 80 47 44 80 50 44 80 50 44 80 50 44 50 50 44 44 47 50 44 80

.59** .73** .74** .56** −.44** −.32* −.46** −.55** −.53** −.56** −.64** −.63** −.48** −.70** −.63** −.33** .45** .65** .65** .30* .29* .07 .38**

1 2 3 4 1 3 4 2 3 4 2 3 4 2 3 2 2 3 3 1 2 3 4

Self-Presentational Efficacy

Body Areas Satisfaction

Appearance Evaluation

Body Image States Fitness Evaluation Self-Classified Weight Physical Appearance State Anxiety-Weight Body Image Avoidance Body Mass Index

* **

p < .05. p ≤ 001.

six weeks). Study volunteers were screened for eligibility via telephone or email. Those who were eligible were then asked to choose an experimental session time from a list of available dates. Using a random numbers table, the experimental sessions were subsequently designated as “same-sex” (n = 23) or “mixed-sex” (n = 24) sessions. Six women were scheduled to participate in each session. Participants were instructed to come to the lab dressed in running shoes, loose-fitting shorts or pants, and a t-shirt. Upon arrival at the lab, participants were seated in a private workspace where they provided informed consent and completed a packet of questionnaires that included a demographics information sheet, and the trait and S-SPAS. Next, the female investigator led participants to a nearby exercise research facility where they were introduced to a certified female fitness instructor who led them through a 45-min strength-training class. The class consisted of 9 exercises performed at 9 different stations. To create the same-sex environment, female confederates exercised on stationary bikes lined up along one wall of the gym, facing the area where the strength-training class took place. For the mixed-sex environment, male confederates exercised on these same stationary bikes. In both conditions, the participants and confederates were in clear view of each other. The participant to confederate ratio was held constant at 2:1 and the absolute number of confederates was adjusted if participants failed to show for their experimental session. In order to control for potential interaction between confederates and participants across the various experimental sessions, the confederates did not participate in the strength-training session, but simply observed the strength-training class from their vantage point on the stationary bikes. They remained on the bikes until the study participants left the exercise facility. After the strength-training session, participants again completed the trait and S-SPAS. Results Reliability of measurement Cronbach’s alpha coefficients indicated acceptable internal consistency reliability (i.e., ˛ > .70) for the S-SPAS across all four

55

samples. In Sample 1, alpha coefficients were .91 at pretest and .90 at posttest. Reliability coefficients were .93, .85, and .70 for Samples 2, 3, and 4, respectively. It is noteworthy that the lowest reliability coefficient (˛ = .70, Sample 4) was observed when participants responded to the S-SPAS by imagining themselves in a particular environment (i.e., a scenario methodology). Reliability coefficients were also acceptable (˛ > .70) for all other scales administered for the purposes of assessing convergent validity, except for the PASTAS. Specifically, the non-weight-related aspects subscale demonstrated inadequate internal consistency (˛ = .52) and was excluded from analyses. Convergent validity Pearson correlation coefficients (two-tailed) were computed to examine the relationship between S-SPAS scores and the measures of body image, self-presentation, and BMI. As shown in Table 2, the S-SPAS was significantly correlated in the expected direction with all of the body image and self-presentation measures. Specifically, higher S-SPAS scores were associated with lower scores for the following scales: Body Areas Satisfaction, Appearance Evaluation, Fitness Evaluation, Body Image States, and Self-Presentational Efficacy (rs ranged from −.32 to −.70, all ps < .05). S-SPAS scores were positively associated with scores on the trait SPAS, Self-Classified Weight, Physical Appearance State Anxiety Scale (weight-related attributes subscale), and the Body Image Avoidance Scale (rs ranged from .56 to .74, all ps < .05). In three out of four samples, there was a significant positive association between S-SPAS scores and BMI (rs ranged from .29 to .38, all ps < .05). Discriminative validity and responsiveness In Sample 1, separate one-way General Linear Model (GLM) Analyses of Covariance (ANCOVA) were conducted on the postexercise state and trait SPA measures. Pre-exercise scores were used as a covariate in each analysis. With regard to the discriminative validity of the S-SPAS, as hypothesized, women in the mixed-sex condition had significantly higher baseline adjusted post-exercise S-SPAS scores (M = 24.63, SE = 1.08) than women in the same-sex condition (M = 19.87, SE = 1.10), F(1, 47) = 9.13, p = .004, d = .92. In addition to demonstrating discriminative validity, this large, between-condition difference in S-SPAS scores also attests to the responsiveness of the S-SPAS. By comparison, baseline adjusted, post-exercise trait SPAS scores were virtually identical for the mixed-sex (M = 28.70, SE = .75) and same-sex conditions (M = 28.56, SE = .76), F(1, 47) = .02, p = .90, d = .04. These results suggest that the trait SPAS is not responsive to situational differences in social physique anxiety. Discussion This study examined the construct validity of a state version of the Social Physique Anxiety Scale (SPAS; Hart et al., 1989) in four different samples. Scores on the state SPAS (S-SPAS) were significantly correlated, in expected directions, with scores on both trait and state measures of body image and self-presentation. There was also some evidence of a relationship between S-SPAS scores and BMI. In addition, S-SPAS scores, but not trait SPAS scores, were significantly higher among women who exercised in a mixed-sex versus a same-sex environment. These findings provide evidence of the validity of the S-SPAS as a measure of situational social physique anxiety. Trait SPAS scores were highly correlated with S-SPAS scores (rs ranged from .56 to .74). This finding is not surprising, as it indicates that women with the greatest dispositional tendencies to experience social physique anxiety did indeed experience the

56

K.A. Martin Ginis et al. / Body Image 8 (2011) 52–57

greatest levels of physique anxiety in experimental situations. A situation-specific measure of self-presentation cognitions – selfpresentational efficacy – was moderately correlated with S-SPAS scores (rs ranged from −.32 to −.46). Women who experienced higher situational social physique anxiety had the lowest confidence in their abilities to present themselves as fit and competent exercisers. This pattern is consistent with Social Cognitive Theory (Bandura, 1986, 1997), which postulates that affective responses to a situation (i.e., situational social physique anxiety) are a determinant of self-efficacy in that situation. In demonstrating this relationship, our study also provides support for the notion that self-presentational variables such as social physique anxiety and self-presentational efficacy, can be examined within the framework of Social Cognitive Theory (Martin Ginis et al., 2007). As predicted, S-SPAS scores were significantly correlated with scores on the body image measures. With regard to the trait body image measures, women who were more satisfied with their physical appearance or who rated their appearance more favorably (i.e., higher scores on the BASS and AE) reported lower state social physique anxiety. It makes sense that women who had more positive thoughts about their appearance, would worry less about how others evaluated their bodies. Likewise, participants who rated themselves as more fit, weighing less (i.e., higher scores on Fitness Evaluation and lower scores on Self-Classified Weight), or less likely to avoid appearance-salient situations (i.e., lower scores on Body Image Avoidance) also had lower S-SPAS scores. Given that all of the samples completed the study measures in exercise-related settings and while wearing exercise clothes, concerns about fitness, body weight, and body-revealing environments would have been salient. As expected, such dispositional concerns were related to the social physique anxiety experienced during the experiment. With regard to the state body image measures, women with higher S-SPAS scores had more negative thoughts and feelings towards their bodies (i.e., lower Body Image States Scale [BISS] score) and reported greater anxiety with regard to weight-related aspects of their physical appearance (i.e., higher Physical Appearance State Anxiety Scale [PASTAS] weight subscale score). Although the correlations between the S-SPAS, BISS and PASTAS were strong, even the largest correlation (r = −.70 between BISS and SPAS) indicated less than 50% common variance between the measures. Thus, the S-SPAS appears to tap into situational feelings about the body that are not fully captured by other state body image measures. As would be expected, there was also some evidence that women with a higher BMI experienced greater state social physique anxiety during the experimental manipulations. In three of the four samples, there were modest, albeit significant positive correlations between BMI and S-SPAS scores. The correlation was not significant for Sample 3, which consisted of women with high levels of trait social physique anxiety (i.e., all participants scored ≥27 on the 9-item SPAS). Given that these women already had a strong dispositional tendency to experience social physique anxiety, it is likely that the highly threatening nature of the experimental manipulation – a body fat analysis by way of a skinfold caliper test – elicited feelings of social physique anxiety regardless of BMI. In addition to demonstrating the S-SPA scale’s validity through concurrent validation, we also demonstrated discriminative validation and responsiveness through our experimental manipulation. Consistent with the findings of Kruisselbrink et al.’s (2004) scenario study, women who exercised in a mixed-sex environment reported significantly higher S-SPAS scores than those who exercised in a female-only environment. These findings are also consistent with the broader social anxiety and self-presentation literature showing that encounters with the other sex are anxiety arousing because of the social awards at stake in such encounters (e.g., the possibility of obtaining feedback that one is sexually desirable, or acquiring a romantic partner; Leary & Kowalski, 1995). Interestingly, whereas

a very large effect (almost a full standard deviation) was found for differences between the two conditions on S-SPAS scores, there was virtually no difference in trait SPAS scores. This finding indicates that, unlike the S-SPAS, the trait version of the SPAS scale is not responsive to situational variations in social physique anxiety. Thus, even though scores on the two versions of the scale are correlated, a state version is needed to detect situational fluctuations in social physique anxiety. Taken together, our study findings support the validity of a state version of the SPAS, and have important implications for the conceptualization, measurement, and further study of social physique anxiety. From a conceptual perspective, our findings add to the accumulating body of evidence indicating that social physique anxiety can be conceptualized as a state that varies across situations (Amirthavasar & Bray, 2007; Kruisselbrink et al., 2004; Martin Ginis et al., 2007). Regarding measurement, the results indicate that state social physique anxiety should be measured with a situation-specific measure, rather than the original, trait SPAS. We recommend that investigators use the version presented in Appendix A. We also recommend further study of state social physique anxiety by investigators working to understand and predict behaviour in physique-salient contexts (e.g., exercise classes, medical examinations, physical education classes), particularly those involving exercise. Indeed, the SPAS has been used extensively in studies of predictors of physical activity and yet a recent meta-analysis showed no statistically significant relationship between SPAS scores and exercise participation (Mack, Wilson, Waddell, & Gasparotto, 2008). We suspect that a state measure of the social physique anxiety experienced in an exercise class or program could be more predictive of adherence to that class or program than a measure of one’s general tendency to experience social physique anxiety (i.e., trait SPAS). If state social physique anxiety is shown to be a significant determinant of exercise, then research could be undertaken to identify modifiable contextual factors that elicit such anxiety, and exercise programs could be developed to alleviate the anxiety. Despite the important contributions of our study, a few limitations warrant mention. First, different study samples completed different measures for the convergent validation analyses. They also completed the questionnaires under different circumstances and using slightly different versions of the S-SPAS. On the one hand, such disparities make it difficult to compare or draw conclusions across studies regarding the scale’s psychometric properties. On the other hand, the different methodologies reflect the different types of studies (e.g., actual versus scenario/imagined manipulations) in which state social physique anxiety has been measured. It is important to determine whether the scale’s psychometric properties are adequate across different types of experimental designs. It is noteworthy that for measures completed by three or more samples (i.e., Appearance Evaluation, Body Areas Satisfaction Scale, trait SPAS, Self-Presentational Efficacy), correlations with the S-SPAS were remarkably consistent across samples, a finding that speaks to the validity of the scale across different experimental conditions. This finding also indicates that rewording items, or the instructions, to make them situationally relevant does not compromise the scale’s internal consistency or construct validity. Appendix A provides a generic wording of the items and the instruction set; however, investigators should feel comfortable making minor modifications to these in order to suit a particular experimental context. A second limitation is that, with the exception of Sample 4, all of the samples were recruited from university campuses (students and staff). Sample 4 consisted of university students as well as young women from neighbouring communities. As such, our study samples are not necessarily representative of women from lower socioeconomic strata or those who are not university educated. Also, because we sampled women only, we cannot make any claims

K.A. Martin Ginis et al. / Body Image 8 (2011) 52–57

regarding the validity of the state SPAS when used in studies of men. Finally, it should be noted that the construct validation of a measure is an ongoing process that involves multiple approaches; no single experiment can unequivocally prove the validity of an instrument (Streiner & Norman, 2008). Additional studies are needed to examine the construct validity of the S-SPAS using other validation approaches. In summary, the present study provides preliminary evidence of the construct validity of a state version of the SPAS. We encourage researchers to begin using the S-SPAS to examine situation-specific expressions of social physique anxiety, and their relationship with exercise and other behaviours that could provoke physique anxiety. Further study of state SPA could lead to a better understanding of people’s behaviour in physique-salient situations and the development of interventions to alleviate situational social physique anxiety. Appendix A. Instruction set and items on the 9-item original (trait) social physique anxiety scale and items on the state social physique anxiety scale. Item

1 2

3

4

5a 6

7

8a

9

Original wording

Revised wording for state version

“Indicate the degree to which the statement is characteristic or true of you” I wish I wasn’t so uptight about my physique/figure. There are times when I am bothered by thoughts that other people are evaluating my weight or muscular development negatively. Unattractive features of my physique/figure make me nervous in certain social settings. In the presence of others, I feel apprehensive about my physique/figure. I am comfortable with how fit my body appears to others. It would make me uncomfortable to know others were evaluating my physique/figure. When it comes to displaying my physique/figure to others, I am a shy person. I usually feel relaxed when it is obvious that others are looking at my physique/figure.

“Indicate the extent to which you are experiencing the feelings described by each item.” I feel uptight about my physique/figure. I am concerned that other people in the room are evaluating my weight or muscular development negatively. Unattractive features of my physique/figure make me nervous.

When in a bathing suit, I often feel nervous about the shape of my body.

In this setting, I feel apprehensive about my physique/figure. I feel comfortable with how fit my body appears to others. It makes me uncomfortable to know that other people are evaluating my physique/figure. When it comes to displaying my physique/figure in this setting, I feel shy. I feel relaxed when it is obvious that other people in the room are looking at my physique/figure. Wearing these clothes, I feel nervous about the shape of my body.

Note. Original wording is taken from Hart et al. (1989). a Reverse-scored items.

57

References Amirthavasar, G., & Bray, S. R. (2007). Women exercisers’ affective reactions to gender variations in class leaders and co-exercisers. Journal of Sport & Exercise Psychology, 30, S146. Bandura, A. (1986). Social foundations of human behavior. New York: Harper & Row, Publishers. Bandura, A. (1997). Self-efficacy: The exercise of control. New York: Freeman. Bane, S., & McAuley, E. (1998). Body image and exercise. In J. L. Duda (Ed.), Advances in sport and exercise psychology measurement (pp. 311–322). Morgantown, WV: Fitness Information Technology. Brown, T. A., Cash, T. F., & Mikulka, P. J. (1990). Attitudinal body image assessment: Factor analysis of the Body-Self Relations Questionnaire. Journal of Personality Assessment, 55, 135–144. Burke, S. M. (2002). The ideal fitness setting: Does the exercise environment affect social physique anxiety? Undergraduate thesis. Hamilton, ON, Canada: McMaster University. Cash, T. F. (2000). Manual for Multidimensional Body-Self Relations Questionnaire. Available from www.body-images.com. Cash, T. F., Fleming, E. C., Alindogan, J., Steadman, L., & Whitehead, A. (2002). Beyond body image as a trait: The development and validation of the Body Image States Scale. Eating Disorders, 10, 103–113. Gammage, K. L., Hall, C. R., & Martin Ginis, K. A. (2004). Self-presentation in exercise contexts: Differences between high and low frequency exercisers. Journal of Applied Social Psychology, 34, 1638–1651. Haimovitz, D., Lansky, L. M., & Oreilly, P. (1993). Fluctuations in body satisfaction across situations. International Journal of Eating Disorders, 13, 77–84. Hart, E. A., Leary, M. R., & Rejeski, W. J. (1989). The measurement of social physique anxiety. Journal of Sport & Exercise Psychology, 11, 94–104. Hausenblas, H. A., Brewer, B. W., & Van Raalte, J. L. (2004). Self-presentation and exercise. Journal of Applied Sport Psychology, 16, 3–18. Kruisselbrink, L. D., Dodge, A. M., Swanburg, S. L., & MacLeod, A. L. (2004). Influence of same-sex and mixed-sex exercise settings on the social physique anxiety and exercise intentions of males and females. Journal of Sport & Exercise Psychology, 26, 616–622. Leary, M. R., & Kowalski, R. M. (1990). Impression management: A literature review and two-component model. Psychological Bulletin, 107, 34–47. Leary, M. R., & Kowalski, R. M. (1995). Social anxiety. New York: Guilford Press. Mack, D. E., Wilson, P. M., Waddell, L., & Gasparotto, J. (2008). Social physique anxiety across physical activity contexts: A meta-analytic review. In J. N. Fuchs (Ed.), Eating disorders in adult women (pp. 1–17). Hauppauge, NY: Nova Science. Marquez, D. X., & McAuley, E. (2001). Physique anxiety and self-efficacy influences on perceptions of physical evaluation. Social Behavior and Personality, 29, 649–659. Martin, K. A., Rejeski, W. J., Leary, M. R., McAuley, E., & Bane, S. (1997). Is the social physique anxiety scale really multidimensional? Conceptual and statistical arguments for a unidimensional model. Journal of Sport & Exercise Psychology, 19, 359–367. Martin Ginis, K. A., & Leary, M. R. (2004). Self-presentational processes in health damaging behavior. Journal of Applied Sport Psychology, 16, 59–74. Martin Ginis, K. A., Lindwall, M., & Prapavessis, H. (2007). Who cares what other people think? Self-presentation in exercise and sport. In G. Tenenbaum & R. C. Eklund (Eds.), Handbook of sport psychology (3rd ed., pp. 136–157). Hoboken, NJ: Wiley. Martin Ginis, K. A., Prapavessis, H., & Haase, A. M. (2008). The effects of physiquesalient and physique non-salient exercise videos on women’s body image, selfpresentational concerns, and exercise motivation. Body Image, 5, 164–172. Martin Ginis, K. A., Strong, H., Arent, S. M., & Bray, S. R. (under review). The effects of acute physique evaluative threat on cortisol activity. Body Image. McAuley, E., Bane, S. M., & Mihalko, S. L. (1995). Exercise in middle-aged adults: Selfefficacy and self-presentational outcomes. Preventive Medicine, 24, 319–328. Reed, D. L., Thompson, J. K., Brannick, M. T., & Sacco, W. P. (1991). Development and validation of the Physical Appearance State and Trait Anxiety Scale (PASTAS). Journal of Anxiety Disorders, 5, 323–332. Rosen, J. C., Srebnik, D., Saltzberg, E., & Wendt, S. (1991). Development of a body image avoidance questionnaire. Psychological Assessment, 3, 32–37. Schlenker, B. R. (1980). Impression management: The self-concept, social identity, and interpersonal relations. Monterey, CA: Brooks/Cole. Streiner, D. L., & Norman, G. R. (2008). Health measurement scales: A practical guide to their development and use (4th ed.). New York, NY: Oxford University Press Inc. Strong, H. (2010). Understanding and changing social physique anxiety among women: The role of cortisol and exercise. Doctoral thesis. Hamilton, ON, Canada: McMaster University. Wasilenko, K. A., Kulik, J. A., & Wanic, R. A. (2007). Effects of social comparisons with peers on women’s body satisfaction and exercise behavior. International Journal of Eating Disorders, 40, 740–745.