A Meta-Analysis Synthesizing 20 Years of Evidence on the Balloon Analogue Risk Task (BART)

While psychological science suffers from measurement proliferation and fragmentation (Anvari et al., 2025), some measures prosper: With over 3,000 citations since its publication approximately 20 years ago, the Balloon Analogue Risk Task (BART; Lejuez et al., 2002) has become one of the most widely used measurement instruments of the behavioral and clinical sciences. Yet, despite its widespread adoption still little is known regarding the task’s psychometric properties in terms of meta-analytic evidence—which may be related to researcher degrees of freedom and flexibility during task implementation that hinder cumulative scientific progress. We thus conducted a robust Bayesian meta-analysis synthesizing 1,667 effect sizes from 220 studies comprising the data of 51,932 participants, to (i) take stock of the psychometric properties of the BART and (ii) quantify measurement flexibility and its impact on the task’s psychometric properties. Although the BART showed satisfactory levels of test–retest reliability (r = 0.54), there was low to no known–groups validity (d = 0.10), convergent validity with other risk-taking measures (r = 0.00), and external validity (r = 0.02). The task’s convergent validity with measures of impulsivity and sensation seeking (r = 0.07) and its divergent validity with measures of cognitive ability (r = 0.12) were also low. There was substantial variability in study characteristics that had an impact on the estimates of all psychometric properties. We make recommendations for future research on the BART and its use, and how psychological science more generally may profit from adopting similar approaches to become a more integrative and cumulative field.