Development and validation of the sports supplements beliefs scale [Conference paper abstract]

Abstract

The World Anti-Doping Agency encourages social science research to uncover the psychological characteristics that render certain athletes susceptible to doping behaviours. Research suggests that athletes using sport supplements are more likely to progress to doping substances and that an athlete’s belief mediates this relationship. However, instruments aiming to measure and quantify athletes’ beliefs have not received appropriate psychometric analysis. This makes it difficult for researchers wishing to measure beliefs and ascertain the accuracy of their results. Therefore, this study aimed to address these concerns via the development and validation of an instrument that measures athletes’ beliefs towards sport supplements. After experts within the field of anti-doping and a select group of athletes reviewed the scales content validity, 171 athletes completed an initial version of the Sport Supplement Belief Scale (SBS). The SBS includes belief statements about the efficacy and use of sport supplements, where participants respond on a Likert-type scale ranging from strongly disagree (1) to strongly agree (6). Theoretically, higher scores on the SBS indicate greater beliefs towards sport supplements. Confirmatory factor analysis (CFA) and the Maximum likelihood procedure hypothesised the explained variance between items and determined acceptable model fit. Model fit was considered acceptable based upon Hu and Bentler’s criteria (1999, Structural Equation Modeling: a Multidisciplinary Journal, 6, 1-55). All participants gave consent to participate in this study and ethical approval was granted by Canterbury Christ Church University. Results of the CFA revealed acceptable model fit for an 11-item version of the SBS (χ2/df = 1.441, RMSEA = 0.050, 90% CI = 0.016 to 0.078, P = 0.452, SRMR = 0.0431, CFI = 0.976, TLI = 0.969). Internal consistency was considered good (α = 0.84) and temporal stability showed very large correlations between one week of first administration (r = 0.86, P < 0.001). Higher scores on the SBS were reported for users of sport supplements compared to non-users (38.6 ± 9.96 vs. 31.0 ± 10.6, P = 0.001, d = 0.8), supporting the scales discriminant validity. The results provide evidence that the SBS is a valid and reliable instrument for measuring athletes’ beliefs towards sport supplements. Future research should aim to incorporate the SBS within the battery of instruments social scientists use to assess athletes susceptible of doping behaviour.