Construct Validity Analysis: Assessment for Learning Primary Mathematics Questionnaire

Abstract

Construct validity is an essential aspect of developing an instrument that involves the collection of evidence to justify decisions. A pilot study was conducted in Hulu Langat, Selangor, to identify the construct validity of a questionnaire developed to measure the practice of Assessment for Learning among primary school Mathematics teachers in the Central Zone, Malaysia. This study applied a simple random sampling technique involving 406 teachers who agreed to perform self-assessment using an online questionnaire. Data analysis was performed using AMOS 23 software. The findings showed only 55 out of 72 items achieved the fit for Confirmatory Factor Analysis. Next, the 12 constructs studied met the statistical requirements for Discriminant Validity and Composite Reliability. The findings of this study are discussed, and suggestions for improvement are given. The construction of a valid and reliable instrument contributes to the study of measurement and evaluation. This contribution meets the needs and fills the research gap in psychometrics.