Psychometric properties of the Turkish version of the comprehensive emotional eating scale

Emotional eating is a multidimensional construct encompassing overeating and undereating in response to positive and negative emotions. The Comprehensive Emotional Eating Scale (CEES) was recently developed to evaluate these four dimensions. However, there is currently no validated Turkish instrument assessing emotional eating in this comprehensive framework. This methodological study aimed to translate the CEES into Turkish and evaluate its psychometric properties in a large adult sample. This study included 897 healthy adults aged 18–65 years. The participants were divided into two demographically matched subsamples for exploratory factor analysis (EFA; n = 449) and confirmatory factor analysis (CFA; n = 448). Construct validity was assessed using EFA and competing CFA models, including CFA, exploratory structural equation modeling (ESEM), and bifactor-ESEM. Reliability was evaluated using Cronbach’s α, composite reliability, and McDonald’s ω. Furthermore, measurement invariance was tested across the categories of gender, age, marital status, and body mass index (BMI). Convergent validity was evaluated through correlations with EDEQ-13, DASS-8, DERS-16, and PSS-14. EFA supported a four-factor structure (positive overeating, positive undereating, negative overeating, and negative undereating) accounting for 70.0% of the total variance, with excellent internal consistency across the subscales (α = 0.948–0.962). The scale also demonstrated good two-week test–retest reliability (ICC = 0.80–0.84 for the subscales; 0.91 for the total score). The CFA results indicated that a bifactor exploratory structural equation model incorporating three specific factors for semantically overlapping items demonstrated the best fit (CFI = 0.996, TLI = 0.994, RMSEA = 0.089, SRMR = 0.043), supporting four subscale factors alongside a weak general factor. In subscale-level invariance analyses, changes in fit indices were generally consistent with metric and scalar invariance across sex, marital status, and age, although the limited absolute fit of the baseline models and the BMI comparisons warrant caution. The CEES scores exhibited theoretically consistent associations with eating disorder psychopathology, emotional dysregulation, stress, and psychological distress. CEES is a valid and reliable measurement tool for the comprehensive and multidimensional assessment of emotional eating behaviors in the Turkish adult population. Its use could enhance research and the multidimensional assessment of emotional eating within Turkish populations, although its clinical utility requires further validation in clinical samples. Emotional eating means changing how much we eat because of how we feel. Most people think of it as eating more when upset, but it is broader than that: some people eat more and others eat less, and this can happen in response to happy feelings as well as difficult ones. A questionnaire called the Comprehensive Emotional Eating Scale (CEES) was recently created to measure all four of these patterns. Until now, there has been no Turkish version of a questionnaire that covers emotional eating this fully. In this study, we translated the CEES into Turkish and checked whether it works well. We asked 897 adults between 18 and 65 years old to complete it, along with other questionnaires about eating problems, emotions, and stress. A group of participants filled it in a second time two weeks later, so we could see whether their answers stayed consistent. The Turkish version measured the four expected patterns of emotional eating, and people's answers were consistent both within the questionnaire and across the two time points. The questionnaire worked in a similar way for women and men, for different age groups, and for married and unmarried people. Scores were also related to eating problems, difficulty managing emotions, and stress, as we expected. This means researchers in Türkiye can now use the CEES to study emotional eating in a more complete way. Further work is needed before it is used with patients in clinics.