The objectives were to conduct a meta-analysis in accordance with Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) standards to determine effect sizes (Cohen's d) for cognitive dysfunction in adults with type 2 diabetes, relative to nondiabetic controls, and to obtain effect sizes for the most commonly reported neuropsychological tests within domains. Twenty-four studies, totaling 26,137 patients (n = 3351 with diabetes), met study inclusion criteria. Small to moderate effect sizes were obtained for five of six domains: motor function (3 studies, n = 2374; d = −0.36), executive function (12 studies, n = 1784; d = −0.33), processing speed (16 studies, n = 3076; d = −0.33), verbal memory (15 studies, n = 4,608; d = −0.28), and visual memory (6 studies, n = 1754; d = −0.26). Effect size was smallest for attention/concentration (14 studies, n = 23,143; d = −0.19). The following tests demonstrated the most notable performance decrements in diabetes samples: Grooved Pegboard (dominant hand) (d = −0.60), Rey Auditory Verbal Learning Test (immediate) (d = −0.40), Trails B (d = −0.39), Rey-Osterreith Complex Figure (delayed) (d = −0.38), Trails A (d = −0.34), and Stroop Part I (d = −0.28). This study provides effect sizes to power future epidemiological and clinical diabetes research studies examining cognitive function and to help inform the selection of neuropsychological tests. (JINS, 2014, 20, 1–14)