Development and validation of a self-efficacy measure for fat intake behaviors of low-income women

Objective: To develop a valid self-efficacy measure for eating low-fat diets in low-income women. Design: Cognitive interviews followed by self-administered questionnaires. Participants/Settings: 206 (6 + 200) nonpregnant black and white women participating in Head Start and the Special Supplemental Nutrition Program for Women, infants and Children (WIC) in Wisconsin. Variables Measured: A 13-item self-efficacy measure for eating low-fat diets. Analysis: Content analysis and factor analyses. Results: After cognitive interviews, 3 items were removed from a draft instrument. Factor analyses were carried out using data from 200 questionnaires that were randomly selected from a pool of 487 respondents: 100 for exploratory factor analysis (EFA) and 100 for confirmatory factor analysis (CFA). Based on EFA, the 10-item pool was further reduced to 8 items, and 3 domains emerged: negative affect, positive affect, and availability. The results of CFA showed excellent model fit to the data (chi(2)/df = 1.2, non-normed fit index 1.0, comparative fit index = 1.0, incremental fit index 1.0, root mean square error of approximate = .046). This measure demonstrated excellent reliabilities (Cronbach alpha: .82-92; composite reliability: .92-99) and construct and convergent validities. Conclusions and Implications: Information obtained from this revised instrument can, be used among low-income women to measure self-efficacy for eating low-fat diets. This is an. important step in developing more targeted and thus more effective education messages to help low-income women modify their fat intake behaviors.