Rasch Measurement Model for validity and reliability of instrument food skills competencies in addressing food neophobia for preschool teachers

Abstract

The aim of this study was to assess the validity and reliability of a newly developed instrument designed to measure food skills competencies in addressing food neophobia (FSFN) among preschool teachers. The Rasch Model Measurement was employed to analyze the validity and reliability of a159-items instrument that designed to measure FSFN of teachers working in one of Malaysian public early childhood institutions. The study included 31 teachers and assistant teachers who were selected through stratified cluster random sampling. The validity of the FSFN instrument was evaluated through four main functional analysis. Overall, both the respondents’ and items’ reliability values in the FSFN instrument were considered excellent and acceptable, ranging from 0.94 to 0.99 and 0.80 to 0.89, respectively. Point-measure correlation analysis revealed no negative values, and item matching analysis resulted to the elimination or revision of 22 items that did not meet the requirements. Additionally, based on the local dependence analysis, the standardized residual correlation value, detected 14 items that required removal. Thus, to enhance the FSFN instrument’s functionality, a total of 36 items were removed, and 123 items remained. This study believed that by eliminating these items, the FSFN instrument’s validity and reliability for the actual study would improve