Developing an instrument to measure physical activity related self-worth in women: Rasch analysis of the Women's Physical Activity Self-Worth Inventory (WPASWI)

ObjectivesThe objective of this study is to report on the development of an instrument to assess non-physical aspects of physical activity (PA)-related self-worth (SW).MethodsThree hundred thirty five women (mean age = 36.69 ± 15.94 yrs, BMI = 24.87 ± 4.56) completed the Women's Physical Activity Self-Worth Inventory (WPASWI), General SW Scale, and a PA Questionnaire. Rasch analysis was used to evaluate the WPASWI.ResultsThree PA-related SW subscales were identified: PA Knowledge (16 items), PA Emotional (13 items), and PA Social (8 items). Rasch analysis supported construct validity, and items demonstrated acceptable internal consistency (Cronbach's alpha = 0.90, 0.87 and 0.72) and test-re-test reliability (r = 0.79, 0.70, 0.81). Women who reported regular PA participation had higher PA Knowledge, PA Emotional, PA Social, and General SW than those who reported some or no PA (p < 0.01) and correlations between Knowledge, Emotional, and Social SW subscales and General SW were 0.207, 0.130, and 0.220, respectively. PA Knowledge and PA Emotional SW had stronger correlations with PA (r = 0.344, 0.273, respectively) than did General SW (r = 0.133).ConclusionsThe WPASWI demonstrated good internal consistency, reliability, construct validity and a stronger relationship with PA than a General SW instrument.

► WPASWI demonstrated three subscales with good fit items for each.
► The findings are consistent with the intended design of the instrument.
► The findings provide initial support for construct validity of this instrument.
► The three WPASWI subscales demonstrated acceptable internal consistency.
► The WPASWI subscales are reliable instruments.