The Psychometric Properties and Performances of Teacher-Made Tests in Measuring Students’ Academic Performance in Ethiopian Public Universities: Baseline Survey Study
Résumé
Abstract All education levels today require quality tests to infer students' scores in judging the quality of teaching and learning. In this judgment, as many scholars argue, teacher-made tests (TMTs) are significant and more promissory than other tests. Besides their benefits, TMTs are prone to measurement error if they do not satisfy assumptions about the psychometric properties and performance of tests. Thus, this study aimed to survey the psychometric properties and performance of teacher-made tests (TMTs) in measuring students’ academic performance using pieces of evidence from three Ethiopian public universities. The researchers employed the post-positivist research paradigm, quantitative research approach, cross-sectional research design, purposive sampling size technique, CTT as the theoretical framework, and SPSS version 28 software as a statistical analysis tool. Empirical data reveal Ethiopian public universities' TMTs quality is below the expected CTT test item standard qualities [content validity, S-CVI=.65<.83, IC reliability, α=.57<.6-.7, difficulty (p)=.49<.5-.6, Skewness=-.300 and D=-0.01<.2], requiring curriculum developers to design test tables, teachers to demonstrate professional accountability, and course offering departments to have a TMTs validation team. Future studies need to include intervention and longitudinal evidence to clarify the association between TMTs and academic performance.
Citer ce document
Accès au document
Texte intégral en lecture en ligne, réservé aux abonnés SPHAERO et aux membres de l'institution. Se connecter
Voir l'article sur le site de la revueAuteur(s)
Statistiques
Consultations : 1
Téléchargements : 0