Analysis of Item Difficulty, Discrimination, and Reliability of the KNP Freshmen Admission Test: Basis for Test Revision and Enhancement

Abstract Admission examinations remain central to higher education institutions because they are used to screen applicants and support student selection decisions. The purpose of this study was to examine the psychometric quality of the KNP Freshmen Admission Test by analyzing item difficulty, item discrimination, and internal consistency reliability. A descriptive-evaluative design was employed using the dichotomously scored responses of 573 examinees to a 100-item admission examination. Guided by Classical Test Theory, the study computed item difficulty indices, upper-lower discrimination indices, corrected item-total correlations, and Cronbach’s alpha. The findings showed that the test had good internal consistency reliability, with a Cronbach’s alpha of .824. However, item-level analysis revealed that the test was generally difficult, with most items falling within the difficult and very difficult ranges. In addition, a substantial number of items demonstrated poor or negative discrimination, indicating that many items require review. Using institutional decision criteria, 28 items were retained, 31 were marked for revision, and 41 were recommended for rejection. These findings imply that, although the test is reasonably consistent as a whole, its item pool requires systematic improvement to strengthen the quality of admission decisions. The study provides an empirical basis for test revision, item banking, and the long-term enhancement of institutional admission assessment practices. Keywords: admission testing, item analysis, item difficulty, item discrimination, reliability, institutional research