A statistical synthesis of the literature on personal and situational variables that predict doping in physical activity settings / Nikos Ntoumanis, Johan Ng, Vassilis Barkoukis, Susan Backhouse
Executive summary
Doping behaviors are not uncommon in sport or certain types of exercise settings. Over the last two decades, there has been a growing interest in understanding the psychological mechanisms associated with doping behavior. As such, there is a need to summarize the evidence in the literature and identify demographic (e.g., sex, age) and psycho-social (e.g., attitudes, perceived norms) variables that are most strongly related to doping behaviors or intentions to engage in such behaviors. To this end, this report represents the first meta- analytical (Hunter & Schmidt, 2004) review of available studies in the extant literature. The review aimed to collectively determine (i) the predictive factors of doping intentions and doping behavior, and (ii) identify moderator variables. A total of 63 studies, spanning 1990- 2013, containing 63 independent published and unpublished datasets from 18 countries were examined and included in an analysis using odds ratios, Pearson correlations, and Cohen’s d for combining study estimates. We found that factors such as the use of legal supplements, perceived social norms, and positive attitudes towards doping were the strongest positive correlates of doping intentions and behaviors. In contrast, factors such as morality and self-efficacy to refrain from doping had the strongest negative association with both intentions
and doping behaviors.
The effects of potential moderators such as sex, publication status, and study design were also tested. Although different sizes of effect were found across distinct levels of moderators, all such effects were in the expected directions. Given that the Theory of Planned Behavior (Ajzen, 1991; Fishbein & Cappella, 2006) has been the dominant framework in doping behavior research (e.g., Zelli et al. 2010; Lazuras et al. 2010; Goulet et al. 2010; Lucidi et al. 2008;), we used path analysis to test a TPB-based model of doping using the meta-analyzed correlations as input matrix. Our results suggested that attitudes and perceived norms (positively), and self-efficacy to refrain from doping negatively) predicted intentions to dope, and in turn predicted doping behaviors. A direct path from perceived norms to doping behaviors was also found, suggesting that perceptions of others’ behaviors may play a salient role in individuals’ doping behaviors. The findings of this meta-analysis provide an objective and concise synthesis of prior research on the psycho-social variables associated with doping behavior and indicate the stronger predictors of doping use. Overall, they highlight the need for a broader approach to anti-doping education. An approach which moves beyond curricula heavily skewed towards compliance and testing towards one that emphasize the social context of doping behavior via strength- and norm-based activities
