Confirmatory factor analysis: Revising a parenting outcome expectancy scale for sexually related discussions with adolescents

Background: Research has identified the importance of parental communication in adolescents' sexual behavior. A scale developed to measure outcome expectancy related to parents talking to their children about sex was tested previously with mothers of adolescents aged 11-14. This study aims to confirm the reliability and validity of the parental outcome expectancy scale. Methods: A 23-item outcome expectancy for sex-based communication was administered to 212 fathers with sons aged 11-14 and 281 mothers with children aged 6-12 as part of interviews conducted for two different sexual health promotion interventions. Confirmatory factor analysis was used to test a 3 factor model based on the results of the previous exploratory factor analysis. Models were assessed systematically, using multiple fit indices. The best fitting models for each dataset were combined and evaluated in both sets. Results: Initially, a combination of the best fitting models for each dataset resulted in indices that fell below acceptable levels. The model was systematically modified yielding a structure with acceptable indices for both datasets (chi-square to degrees of freedom ratios of 2.26 and 2.60, GFIs of .90 and .91 and RMSEAs of 0.078 and 0.078 in the sample of fathers and mothers, respectively). Internal consistency was acceptable to high across both groups, and correlations between the scale and having talked about sex were significant and positive. Conclusions: The reduced 14-item parental outcome expectancy scale for sexually related discussions demonstrates reliability and validity. Results indicate the scale could be used with fathers and mothers in studies about sex-related communication.