REiD (Research and Evaluation in Education)
Vol. 12 No. 1 (2026)

Psychometric properties of a scale for measuring students’ confidence in mathematics learning using the Partial Credit Model

Raoda Ismail (Cenderawasih University)
Agus Santoso (Universitas Terbuka)
Okky Riswandha Imawan (Universitas Cenderawasih)



Article Info

Publish Date
08 Sep 2026

Abstract

Self-confidence is a crucial affective factor that influences students’ engagement, persistence, and achievement in mathematics learning. This study aimed to examine the psychometric properties of a scale for measuring students’ confidence in mathematics learning using the Partial Credit Model (PCM) within the framework of Item Response Theory (IRT). A total of 273 Indonesian senior high school students were selected through random sampling and responded to a 20-item five-point confidence inventory. Data were analyzed using PCM to examine item fit, threshold parameters, reliability, and test information. The assumptions of unidimensionality and local independence were tested using exploratory factor analysis and Yen’s Q3 statistic. The results indicated that 15 of the 20 items fit the PCM model adequately, while five items (1, 3, 7, 14, and 19) showed misfit and require further revision. The instrument demonstrated good reliability (0.78). The test information function revealed that the instrument measured students’ confidence most precisely within the ability range of −2.90 to 2.32. These findings confirm that the validated instrument provides a reliable and psychometrically sound tool for assessing students’ confidence in mathematics learning. The instrument can be used by educators and researchers to diagnose students’ confidence levels and support the development of targeted instructional interventions.

Copyrights © 2026