A Rasch Analysis of Student Coding Attitudes Survey

Abstract

The purpose of this study was to examine the psychometric properties of a 23-item scale preliminarily entitled the “Elementary Student Coding Attitudes Survey” (ESCAS) from a Rasch perspective. The ESCAS includes five latent constructs to assess attitudes: coding confidence, interest, utility, perceptions of coders, and social value using Item Response Theory [IRT]. The item summary statistics, person summary statistics, item misfit statistics, and category structure statistics were examined for each component. In examining the five constructs separately, the measure had above average psychometric properties with no misfitting items and higher internal-consistency reliability.