Reliability generalization meta-analysis of the food neophobia scale: Turkish sample

Abstract

Purpose: The Food Neophobia Scale developed by Pliner and Hobden (1992) has been widely utilized globally and in Turkey for many years to measure people's fear of new foods. This study aims to determine the mean value by conducting a reliability generalization meta-analysis for the reported reliability coefficients of individual studies in Turkey's tourism field, which employed Pliner and Hobden's scale to investigate food neophobia. Additionally, this study explores variations in the mean value among subgroups. Methods: A reliability generalization meta-analysis based on a random-effects model was conducted to examine the heterogeneity of reliability coefficients in the study, along with heterogeneity analyses and moderator analyses. Results: Based on the analysis of 48 independent samples (N= 23306), the transformed mean Cronbach's alpha value was estimated to be .827 (95% CI [.796-.853]) and found to be significant. The Q-test and I2 values reveal significant heterogeneity between alpha coefficients, indicating a notable variation in the measurement reliability across samples. Moderator analyses using analog to the ANOVA and meta-regression analyses showed that reliability coefficients differed according to the variables of publication type, sample type, and proportion of women in the sample. Implications: The results offer valuable insight for researchers seeking to select appropriate scales for investigating food neophobia.