Journal of Public Health | 2019
One year reliability of the Dutch eating behavior questionnaire: an extension into clinical population
Abstract
Previous research introduced the Dutch Eating Behavior Questionnaire that allows the measurement of emotional, external and restrained eating styles. However, only limited data exist about long-term test-retest reliability, particularly for clinical participants. The current study evaluated the Dutch Eating Behavior Questionnaire 12-month test-retest reliability and internal consistency using the responses of 494 participants (241 pre-bariatric participants and 253 normal-weight controls). Internal consistency remained acceptable for each group at both time points. Cronbach’s α ranged from 0.81 (restrained eating at time 1 in pre-bariatric group) to 0.94 (external eating at time 1 in normal-weight group). Also, the stability estimates were satisfactory and ranged from 0.79 (restrained eating in pre-bariatric group) to 0.92 (external eating in normal-weight group). It is concluded that the internal consistency of the Dutch Eating Behavior Questionnaire is excellent, and that the test–retest reliability results suggest that items intended to measure three types of eating styles were likely testing trait, rather than state, factors, regardless of whether normal-weight controls were included in, or excluded from, the analyses. Finally, the measurement invariance of the Dutch Eating Behavior Questionnaire across body mass index status was assessed. It indicated that the three-factor structure had the best fit to the data, and this dimensional structure proved to be invariant across body mass index status.