Since the COVID-19 pandemic, the use of open-book examinations (OBE) in higher education has increased compared to closed-book examinations (CBE), raising questions about the relative efficacy of these two major examination formats. We review the effects of CBE and OBE on learning and examine potential moderating variables related to study, course, and assessment characteristics. Meta-analytic results indicate that the overall effect of examination format is not statistically significant (k = 44, N = 3,499, g = -0.22, 95% CI [-0.75, 0.32]), suggesting that both formats have comparable effectiveness. Results also indicate that the effect of examination format is moderated by the retention interval, with CBE offering greater benefits as learning assessments become increasingly rigorous (g = 0.02). We discuss the implications of these findings and recommend that educators implement more OBE because they alleviate students' test anxiety and better align with Universal Design for Learning principles. Our study provides empirical support to inform the pedagogical decisions of instructors who seek guidance on the merits of CBE and OBE.