Psychometric properties of the Turkish version of the child food rejection scale.

Abstract

Background: While some children are open to trying new foods, many exhibit dislike or refusal when encountering them for the first time. These behaviors can be broadly characterized as food neophobia and pickiness, which are believed to be the primary forms of food rejection among children. Because there are differences between countries in early feeding practices and culinary traditions, culturally adapted scales are needed to measure these behaviors. This study aimed to (1) test the reliability and validity of the Turkish adaptation of the Child Food Rejection Scale (CFRS), and (2) examine patterns of food rejection and the correlations between scale scores and sociodemographic characteristics among Turkish children.

Methods: The validity of the CFRS translated into Turkish and cross-culturally adapted was assessed by content validity, construct validity, and convergent validity. Internal consistency was measured with Cronbach's alpha coefficient for the scale and its subdimensions. The scale reliability was also evaluated using test-retest reliability and several tests.

Results: Three hundred seventy-five primary caregivers (mainly mothers) of children aged 2 to 7 years were recruited through an online questionnaire. The results confirmed the psychometric soundness of the Turkish CFRS. Construct validity was supported by factor analysis (KMO coefficient = 0.852; Bartlett's sphericity test χ²=1301.580, p < 0.01), with two factors explaining 53.47% of the variance. Internal consistency was high (Cronbach's α = 0.838 for the total scale; α = 0.845 for neophobia, α = 0.600 for pickiness). Test-retest reliability (ICC = 0.770) indicated stability over time. Reliability was further supported by split-half reliability (Spearman-Brown coefficient = 0.749; Guttman split-half coefficient = 0.746), and the scale demonstrated additivity (Tukey's test F = 35.543, p ≤ 0.001) and absence of response bias (Hotelling T-square test F = 63.041, p ≤ 0.001). Confirmatory factor analysis showed good model fit for the two-factor structure, supporting construct validity. Moreover, the rate of food rejection was 21.1%, and CFRS scores did not vary by gender, age, or other sociodemographic characteristics among Turkish children.

Conclusions: Overall, these findings affirm that the adapted CFRS is a reliable and valid tool for assessing food rejection behaviors in Turkish children.