Factor analyses of condom attitudes, norms, and self-efficacy measures in diverse samples.

ABSTRACT

Measurement of condom attitudes, norms, and self-efficacy is variable, reducing research generalizability. The purpose of this study was to assess reliability and construct validity of measures of condom attitudes, norms, and self-efficacy across 3,989 youth. Data were collected at computer kiosks and on the Internet. Analyses included internal consistency reliability assessments and exploratory and confirmatory factor analyses of three constructs in a 32-item a priori instrument. Reliability estimates ranged between 0.68 and 0.89, and goodness of fit indices all exceeded 0.90. Factor analysis results supported the compatibility of the items from the three subscales with their conceptual domains. Final scales include measures of positive and negative outcome expectancies related to condom use, peer and partner condom norms, self-efficacy for condom negotiation, and self-efficacy for condom use. These measures of condom attitudes, norms, and self-efficacy can be employed for assessments to improve standardization of measures and generalizability of research, particularly related to HIV prevention.