Self-efficacy is a consistent correlate of physical activity, but most self-efficacy measures have not been validated in diverse populations. This study examined the construct, criterion-related, and convergent validity and internal consistency of the Self-Efficacy for Exercise Questionnaire.
African American and Caucasian adults (N = 1919) from two adjacent counties in South Carolina were identified through a list-assisted random digit-dialed telephone survey. Psychometric properties of the measure were assessed by gender, race, age, education, and body weight subgroups.
Across all subgroups, a single-factor solution explained 93 to 98% of the common variance in an exploratory factor analysis, and all 14 items had factor loadings exceeding 0.40. Higher exercise self-efficacy was significantly associated with greater physical activity, younger age, male gender, higher education, and lower body weight, as predicted. Internal consistency was high for all subgroups (α = 0.90 to 0.94).
The Self-Efficacy for Exercise Questionnaire appears to be a valid and reliable measure for use with diverse populations.
The authors are with the Arnold School of Public Health, University of South Carolina, Columbia, SC 29208. Wilcox and Sharpe are with the school’s Dept of Exercise Science. Sharpe, Hutto, and Granner are also with the school’s Prevention Research Center. Granner is also with the school’s Dept of Health Promotion, Education, and Behavior.