A Brief Measure of Peer Affiliation and Social Acceptance (PASA): Validity in an Ethnically Diverse Sample of Early Adolescents

Abstract

The purpose of this study was to conduct a multiagent–multimethod analysis of the validity of a brief measure of deviant peer affiliations and social acceptance (PASA) in young adolescents. Peer relationships are critical to child and adolescent social and emotional development, but currently available measures are tedious and time consuming. The PASA consists of a youth, parent, and teacher report that can be collected longitudinally to study development and intervention effectiveness. This longitudinal study included 998 middle school students and their families. We collected the PASA and peer sociometrics data in Grade 7 and a multiagent–multimethod construct of deviant peer clustering in Grade 8. Confirmatory factor analyses of the multiagent–multimethod data revealed that the constructs of deviant peer affiliations and social acceptance and rejection were distinguishable as unique but correlated constructs within the PASA. Convergent, discriminant, concurrent, and predictive validity of the PASA was satisfactory, although the acceptance and rejection constructs were highly correlated and showed similar patterns of concurrent validity. Factor invariance was established for mother reports and for father reports. Results suggest that the PASA is a valid and reliable measure of peer affiliation and of social acceptance among peers during the middle school years and provides a comprehensive yet brief assessment of peer affiliations and social acceptance.

Notes

¹Results of analyzing 401 observations were essentially the same as the results with 998 observations.

Note: (R) = reverse-scored indicators.

Note: DPC = deviant peer clustering; PA = peer affiliation; ACC = acceptance; REJ = rejection; (R) = reverse-scored indicators.

Note: CFI = comparative fit index; RMSEA = root mean square error of approximation; CI = confidence interval; SRMR = standardized root mean square residual; PASA = brief measure of peer affiliation and social acceptance.

²A reviewer has questioned whether informants who share different environments can show invariance in measuring the factors. We conducted additional analyses to test whether the teacher and parent, child and parent, and child and teacher reports were invariant. As a result, the metric invariance test (invariance among the factor loadings) indicated child and parent, and teacher and parent were not invariant. The factor loadings for child and teacher ratings were invariant, but the scalar invariance test (invariance between the intercepts) showed that the intercept for child and teacher ratings was noninvariant. In conclusion, only the mother and father reports were invariant in measuring the factors.