Kajian Validitas dan Reliabilitas Tes dalam Evaluasi Pembelajaran

Abstract

This study aims to examine the concepts of validity and reliability as two fundamental aspects that determine the quality of research instruments. Validity refers to the extent to which an instrument measures what it is intended to measure, whereas reliability indicates the consistency of the measurement results obtained. The study employed a library research method by reviewing various relevant scientific sources, including books, journal articles, and academic documents discussing validity and reliability in research instruments. The findings reveal that validity consists of several types, namely content validity, construct validity, criterion validity, face validity, and empirical validity. Meanwhile, reliability includes test-retest reliability, parallel forms reliability, internal consistency reliability, inter-rater reliability, and intra-rater reliability. Each type of validity and reliability possesses distinct characteristics and testing procedures depending on the objectives and forms of the instruments used. Therefore, validity and reliability testing should be conducted systematically to ensure that research instruments produce accurate, consistent, and trustworthy data. This study is expected to serve as a reference for researchers and educational practitioners in understanding the importance of validity and reliability as the foundation for developing high-quality research instruments.