The Analysis of Religious Character Instrument using Classical and Modern Theories

Abstract

Measuring religious character is essential for understanding how individuals internalize religious values. However, previous studies have mainly focused on conceptual development or instrument construction without systematically comparing psychometric characteristics before and after scaling or integrating Classical Test Theory (CTT) and Item Response Theory (IRT) within a single analytical framework. This study addresses these gaps by examining the characteristics of a multidimensional religious character instrument before and after scaling to ensure score stability and measurement precision. The instrument was a self-report questionnaire using 4–5 point Likert-type items representing five dimensions: intellectuality, ideology, public practice, private practice, and religious experience. A descriptive quantitative design was employed. CTT was used to evaluate item statistics, reliability, and score distribution, while IRT specifically the graded response model (GRM) assessed item functioning across different levels of the latent trait. The summated rating method was applied to transform ordinal responses into standardized scores. Data were collected from 375 students at Widya Dharma University, Klaten, Indonesia, and analyzed using R. The scaling procedure generated positivized z-scores and produced a more compressed score distribution, reflected in decreased mean, standard deviation, mode, and median. Changes in reliability coefficients and the Standard Error of Measurement (SEM) across dimensions indicated that scaling affected measurement precision. GRM analysis confirmed that the instrument effectively discriminated among individuals with low, moderate, and high levels of religious character. Overall, the findings highlight the value of applying scaling procedures and integrating CTT and IRT to improve the accuracy, interpretability, and psychometric robustness of religious character assessments.