Measuring Learning Loss Due to Disruptions from COVID-19: Perspectives from the Concept of Fractions

Abstract

Ineffective learning during Covid-19 can be a major cause of learning loss. Through polytomous item response theory analysis, this study seeks to identify learning loss of fractions and investigate the reliability and validity of the DINA model diagnostic test. This study involved 177 Grade 8 students from six junior high schools in Banda Aceh, Indonesia, to obtain information about the validity and reliability of the diagnostic test and learning loss of fraction due to Covid-19. The research was conducted by designing diagnostic test, expert validation, and empirical testing. Multiple choice problems with polytomous options were developed. This study produced 20 valid problems based on the level of item discriminant, item difficulty, and slip and guess, with a reliability of 0.899. This study revealed that learning loss was experienced by 62 students from low-level junior high schools (100%), 51 students from middle-level junior high schools (98%), and 54 students from high-level junior high schools (85.7%). It can be said that low-level schools have the highest percentage of students with learning loss. It is recommended that future studies evaluate the effectiveness of diagnostic tests in identifying learning weaknesses in mathematics before semester exams using the items generated in this study.