ANALISIS DAYA PEMBEDA PENGECOH DAN TINGKAT KESUKARAN SOAL PENDIDIKAN AGAMA ISLAM DALAM UJIAN AHIR SEMSTER TINGKAT SEKOLAH DASAR: STUDI KASUS SDN KALIWINIG 07 RAMBIPUJI JEMBER

Abstract

Abstract

Tests are methods or procedures used in educational measurement and assessment. Tests involve giving tasks or a series of tasks to test takers, either in the form of questions that must be answered or commands that must be carried out, to produce a score that reflects the test taker's behavior or achievements. Question item analysis is an activity to evaluate each item or question item to assess its quality. Item analysis is the process of reviewing questions in a test with the aim of obtaining a set of quality questions. Therefore, analysis is needed regarding the quality of the questions, both in terms of level of difficulty, discrimination and distracting power. The research used to analyze the question items used a quantitative descriptive research design. Type of quantitative descriptive research. The population in this study were fourth grade students at SDN Kaliwining 07 Rambipuji Jember. The research sample consisted of 21 students in one class, with the class IV Islamic Religious Education subject as the research object. The instruments used in this research include end-of-semester exam question sheets, answer keys, and student answer sheets which are used as data for analysis. The data collection techniques used in this research are observation and documentation. Quantitative data analysis in this research was carried out using a computer application called Iteman, which is special software for analyzing test items. The results of this analysis show that the level of difficulty of the question items shows that each question item has a good level of difficulty. This can be seen from the question items which have a difficulty value or P value between 0.3-0.7, which is the level of difficulty stated. good if it has a vulnerability of 0.2-0.8. So that the questions analyzed based on students' answers have a good level of difficulty. Meanwhile, the different power results of the questions get several different results, from bad, average, and good. This is shown by the Rpbis results A. 0.496, B. 0.023, C. -0.166, and D. -0.166. So some of the differentiating powers contained in the question items are not yet able to distinguish between students who differentiate between the abilities of students who have mastered and who have not mastered the competencies achieved. Therefore, improvements are needed from the school or educators. The distracting power, which obtained good and very good criteria, can be seen from the results of prop A. 0.095, B. 0.382, C. 0.429, D. 0.095. which are classified as good and very good criteria, so that each question item is chosen in the same proportion by students.