Analisis parameter tes penilaian akhir semester fisika kelas X dengan teori respon butir

Abstract

The results of a preliminary study conducted through interviews with several physics teachers at one of the high schools in Bandung showed that the quality of the test instrument was analysed only if needed or not carried out regularly so we couldn't know how the quality of the test instruments. Although the analysis was carried out, the analysis used classical analysis. The quality of the test instrument can be seen from the characteristics of the test itself which is obtained by analyzing the parameters items on the instrument. Therefore, a study was conducted that aims to analyse the parameters of the final semester assessment test for Physics class X for the 2020/2021 academic year with item response theory. The method used in this research is descriptive quantitative method. The participants in this study were 215 students of class X MIPA in one of the high schools in Bandung. All of the data were then analysed using the item response theory of the 2 PL model. This model was chosen because it has the highest peak value of information, which is 25.132. The results of the analysis show that the Final Semester Assessment test have 2 items that are not valid from 25 items (5th and 12th items), the value of the discriminating power parameter (a) is 0.354 which is categorized as good, and the value of the level of difficulty parameter (a) is 0.354. b) of -1.4 which is in the easy category. This Final Semester Assessment test has a peak value of the information function of 25.132 and SEM (Standard Error of Measurement) of 0.199 with the intersection between the two curves in the range of -3.619 to 1.206 so that the question is reliable to measure very low to high abilities.