Reliability and Validity in Quantitative Research: A Simplified Guide for Researchers

Résumé fourni par la source

Reliability and validity are the two fundamental properties that determine the quality of measurement in any quantitative study, and their careful evaluation is vital for producing credible and reproducible research. These psychometric properties greatly affect the internal and external validity and the overall trustworthiness of study conclusions. This paper comprehensively updates and simplifies the concepts, types, and estimation of reliability and validity, providing evidence to assist researchers in advancing the rigour and statistical defensibility of their instruments. The principal forms of reliability—test–retest, internal consistency, inter-rater, parallel-forms, and split-half—and the principal forms of validity—content, face, criterion, and construct—are fully explained. Reliability reflects the consistency and stability of measurement, while validity reflects the degree to which an instrument measures what it is intended to measure. The quantification of these properties is addressed through Cronbach's alpha, the Pearson correlation, Cohen's kappa, the intraclass correlation coefficient, and the content validity index, together with recommended interpretation thresholds. The paper contributes both theoretical guidance and practical tools for selecting appropriate estimation strategies and for reporting psychometrically defensible evidence of instrument quality. In sum, it sets a standard for best practice in research methodology that drives reliability, validity, and empirical rigour across diverse studies.

Ce résumé expose les affirmations des auteurs. BNTIC ne l’interprète pas comme une validation indépendante des résultats.