Abstract Individuals with ADHD are characterized as ubiquitously slower and more variable than their unaffected peers, and increased reaction time (RT) variability is considered by many to reflect an etiologically important characteristic of ADHD. The present review critically evaluates these claims through meta-analysis of 319 studies of RT variability in children, adolescents, and adults with ADHD relative to typically developing (TD) groups, clinical control groups, and themselves (subtype comparisons, treatment and motivation effects). Random effects models corrected for measurement unreliability and publication bias revealed that children/adolescents (Hedges' g=0.76) and adults (g=0.46) with ADHD demonstrated greater RT variability relative to TD groups. This increased variability was attenuated by psychostimulant treatment (g=−0.74), but unaffected by non-stimulant medical and psychosocial interventions. Individuals with ADHD did not evince slower processing speed (mean RT) after accounting for RT variability, whereas large magnitude RT variability deficits remained after accounting for mean RT. Adolescents and adults with ADHD were indistinguishable from clinical control groups, and children with ADHD were only minimally more variable than clinical control children (g=0.25). Collectively, results of the meta-analysis indicate that RT variability reflects a stable feature of ADHD and other clinical disorders that is robust to systematic differences across studies.