Umbrella review of meta-analyses on the benefits and risks of computerised adaptive testing

Computerised adaptive testing (CAT) has been widely applied in large-scale assessments worldwide. To create a comprehensive action guide, practical evidence should be synthesised. We conducted a pre-registered umbrella review (PROSPERO CRD420251207547) of meta-analyses on the benefits and risks of CAT, incorporating 10 meta-analyses within six reports (sample size range 2830–32,735, sample size median 16,842; searched via PubMed, ProQuest PsycINFO, Scopus, and Web of Science). By combining the effect sizes and conducting a re-meta-analysis, we found that CAT is beneficial for test-takers’ motivation ( d = 0.08) and positive emotions ( d = 0.13), but it is also associated with increased risk for test-takers’ anxiety ( d = 0.08) and lower test scores (median d = −0.08). A quality appraisal was conducted following the Joanna Briggs Institute Critical Appraisal of Systematic Reviews and Research Synthesis tool. The present review found the simultaneous presence of increased motivation and anxiety associated with the application of CAT, along with the enriched CAT experience made possible by the prevalence of technology. We suggest that more high-quality meta-analyses should be conducted on the benefits and risks of CAT in a broader context, considering a wider range of application scenarios and more diverse outcome variables.