Analysis of the Quality of Final Semester Examination Items in Islamic Religious Education for Grade Vii at Smp N 4 Satu Atap Panai Hulu

Abstract

This study aims to analyze the quality of the Final Semester Examination (UAS) items for the Islamic
Religious Education (PAI) subject in Grade VII at SMP N 4 Satu Atap Panai Hulu based on validity,
reliability, item difficulty, discrimination power, and distractor effectiveness. This study employed a
quantitative approach using a descriptive-evaluative method. The research data were obtained through
documentation of the examination items, test blueprint, answer key, and students' answer sheets, as well
as interviews with the PAI teacher as supporting data. The analysis was conducted through a logical
approach covering the aspects of content, construction, and language, as well as an empirical analysis
based on students' test results. The findings show that, in general, the Grade VII PAI Final Semester
Examination instrument is categorized as reasonably suitable for use as an instrument for assessing
learning outcomes. In terms of content, construction, and language, most of the test items have met the
basic criteria for good test-item construction. However, the empirical analysis indicates that there are
still weaknesses in empirical validity, reliability, item difficulty, discrimination power, and distractor
effectiveness. The majority of the items fall into the easy to moderate difficulty categories, while
relatively few items are classified as difficult. The dominance of items measuring lower cognitive levels
also results in low discrimination power for several items and makes the instrument less effective in
distinguishing students' abilities. Therefore, improvement and development of the test items are
necessary by considering a balance of cognitive levels, item difficulty, discrimination power, and
distractor effectiveness so that the PAI assessment instrument becomes more valid, reliable, objective,
and comprehensive.