LAUSR

INTRODUCTION The purpose of this study was to assess the reliability and construct validity of a measure designed to evaluate self-efficacy to reduce sedentary behavior (SB). METHODS Initial instrument development was based on semi-structured interviews and in-depth review of existing measures of self-efficacy for physical activity (PA). Items were drafted by the study authors and reviewed by SB experts. Participants recruited via Amazon Mechanical Turk completed the pool of items and the Exercise Confidence Survey and provided self-reported PA, SB, and demographic information. The sample was divided into two random halves and exploratory factor analysis (EFA) and confirmatory factor analysis (CFA) were performed on each half, respectively. Internal consistency reliability (Cronbach's alpha) was calculated for the final scale. Initial criterion validity was examined in relation to self-reported SB and PA. Analyses were conducted in SAS 9.4 and Mplus 8.3. RESULTS Data were collected from N = 818 adults (47.6% women, mean (SD) age 37.8 (10.6)). EFA results strongly supported a one-factor scale. Items with low factor loadings (<.65) were dropped from the scale; 10 items were retained. CFA demonstrated that the 10-item measure fit the data well, but one item had a low factor loading. A final scale with nine items was retained, with good fit to the data (χ2 (27) =90.79, p < 0.0001, CFI = 0.97, RMSEA = 0.08 [90% CI = 0.06, 0.09], SRMR = 0.03), and all items had high factor loadings (>0.70). Internal consistency reliability was high (α = 0.91). Self-efficacy to reduce SB was significantly and positively correlated with exercise confidence (r = 0.32-0.38, p < 0.0001). CONCLUSIONS We developed a nine-item measure of self-efficacy to reduce SB that demonstrates strong initial psychometric properties. Although related to exercise self-efficacy, self-efficacy to reduce SB represents a distinct construct.