Cross-Validation Study Using Item Response Theory

Abstract

The Health-Related Quality of Life for Eating Disorder-Short questionnaire is one of the most suitable existing instruments for measuring quality of life in patients with eating disorders. The objective of the study was to evaluate its reliability, validity, and responsiveness in a cohort of 377 patients. A comprehensive validation process was performed, including confirmatory factor analysis and a graded response model, and assessments of reliability and responsiveness at 1 year of follow-up. The confirmatory factor analysis confirmed the two second-order latent traits, social maladjustment, and mental health and functionality. The graded response model results showed that all items were good for discriminating their respective latent traits. Cronbach's alpha coefficients were high, and responsiveness parameters showed moderate changes. In conclusion, this short questionnaire has good psychometric properties. Its simplicity and ease of application further enhance its acceptability and usefulness in clinical research and trials, as well as in routine practice.

Using Item and Test Information to Optimize Targeted Assessments of Psychological Distress

“…Again, factor analysis and DIF tests were applied. The central test to evaluate the fit of the model was the test whether the item parameters from the estimation sample generalize to the validation sample (Bilbao et al, 2013; Hawkins, 2004). This was checked by cross-validating the derived item parameters from the estimation sample on the second, the validation sample, and assessing the fit statistics of the items.…”