Validation of the Eating Attitudes Test-26 in a Mexican female sample: confirmatory factor analysis and cut-off point

The objective of the present study was to determine the psychometric properties of the Eating Attitudes Test (EAT-26) in a sample of Mexican women, report its validity through confirmatory factor analysis, and to determine the cut-off point for the test. The sample consisted of 1088 women, aged between 12 and 27 years ( = 19.31, SD = 2.94), from educational institutions of the metropolitan area of Mexico City and a Clinical Group of 108 patients, aged between 11 and 40 years ( = 19.35, SD = 5.26). Regarding the results, the instrument obtained a Cronbach’s alpha of .85 and .92 in community and clinical sample, respectively. The construct validity yielded four factors that explained 45.71% of the total variance. To corroborate the adequacy of the factor structure, a confirmatory factor analysis was performed, yielding the following validity indicators: GFI = 0.93; CFI = 0.90 and RMSEA = 0.055. Finally, a 0-4 scoring system was used for the Mexican population, as it was considered more appropriate. The cut-off point determined was ≥ 40 with a sensitivity of 0.75, and a specificity of .88. In conclusion, the findings from this study indicate that the EAT-26 shows acceptable reliability, and a meaningful factor in clinical and non-clinical samples of Mexican women.