Abstract
The Eating Attitudes Test 26 (EAT-26) scale is a well-established tool for assessing the risk of eating disorders. A reliability generalization meta-analysis was conducted to estimate the average reliability of the EAT-26 scale scores and how reliability estimates vary according to the composition and variability of samples, to identify study characteristics that can explain its variability, and to estimate the reliability induction rate. A literature search produced 14 articles involving 15 studies that met the inclusion criteria. For the total scores of the EAT-26 scale, pooled Cronbach's alpha was 0.85, with 95% confidence intervals of 0.81 and 0.88, a standard error (SE) of 0.02, and a standard score of (Z = 43.99). Moderator analysis showed that the language, age, or sex of the participants did not affect the overall results. By assessing the reliability of research findings, researchers can examine the consistency of results across studies, which can help identify sources of variability in the results.