Social anxiety disorder in adolescents is increasingly recognized as a common condition that may precede onset of other mental health problems. However, few measures are currently available to screen for adolescent social anxiety, and little is known about their psychometric characteristics in school-based samples. To this end, the present study was undertaken as a psychometric cross validation of the Social Anxiety Scale for Adolescents (SAS-A). Exploratory and Confirmatory Factor Analyses (CFA) supported the original 3-factor structure of the SAS-A but retained fewer items than in the original scale. The revised scales demonstrated good internal consistency as well as criterion and concurrent validity. Despite including fewer items, psychometric characteristics of the revised scales were equivalent to or stronger than those reported in previous studies of the measure. As such, the present study provides initial support for the utility of the revised SAS-A as a measure of social anxiety in adolescent school samples.
Social anxiety among adolescents has received increasing attention in recent years, yet a limited literature addresses this issue (for a review see Chavira & Stein, 2000). Social anxiety disorder in adolescents is being increasingly recognized as a common condition (Wittchen, Stein, & Kessler, 1999), one that often heralds the onset of other mental health problems such as depressive disorders (Pine, Cohen, Gurley, Brook, & Ma, 1998). This suggests that screening efforts directed at identifying adolescents with social anxiety disorder might have merit. But, at present, relatively few measures of adolescent social anxiety are available, and little is known about their psychometric characteristics in school-based samples.
Of the few measures of adolescent social anxiety currently available, the Social Anxiety Scales for Adolescents (SAS-A; La Greca, 1999) as been the most extensively studied. While a number of studies have examined the psychometric characteristics of the original version of these scales (the Social Anxiety Scales) for Children (SAS-C; for a review see La Greca, 1999), only one published study has examined psychometrics of the SAS-A (La Greca & Lopez, 1998), providing support for a 3-factor solution with good internal reliability, test–retest stability, and construct validity. The primary goal of the present study was to examine the psychometric characteristics of the SAS-A when self-administered in a sample of high school students. To this end, a confirmatory factor analytic approach was used to examine the fit of the current data to the previously reported scale structure. The emergent and original scale structures were then submitted to validation analyses. Construct validity was evaluated by examining derived scale scores in relation to endorsement of social anxiety diagnostic criteria. We expected that higher scores on the SAS-A scales would correspond with endorsement of more social anxiety criteria. Concurrent validity was assessed in relation to negative emotionality. Previous psychometric studies have demonstrated a relationship between social anxiety scale scores and measures of negative affect (La Greca, 1998; Schuman, La Greca, Vaughn, Kirby, & Bauer, 1994). As such, we anticipated significant positive correlations between SAS-A scale scores and scores on the Negative Emotionality Scale (NES). Finally, since previous studies have reported gender differences for social anxiety (Inderbitzen and Hope, 1995 and La Greca and Lopez, 1998) we expected that girls would obtain higher score on SAS-A scales.
