The purpose of this study was to assess the reliability and validity of a Dutch translation of a questionnaire to measure self-efficacy in leisure-time physical activity. The questionnaire consisted of three subscales measuring three dimensions of self-efficacy. It was completed by 461 participants, 55–65 years old. Fifty-nine participants took part in a test-retest study. Factor analysis and correlations between the sum-scores of the 3 scales confirmed that each scale measures a different dimension of self-efficacy. The criterion-related validity of 2 of the scales was found to be moderate. All 3 scales had a satisfactory internal consistency, indicating that they are reliable. Stability was assessed with a test-retest procedure, which yielded satisfactory results for 2 of the 3 scales. The results revealed an improvement in self-efficacy for 2 of the scales over a 4-week time period. When outliers were excluded, satisfactory values were obtained for intraclass correlation coefficients between the first and second measurements.