Analisis pola respon untuk mengetahui kemampuan peserta didik dengan pendekatan satu parameter (1PL)

Abstract

Learning outcomes from a learning activity are in the form of changes in the ability of learners to be better. The learning outcomes are in the form of cognitive, affective, social, and sensomotor abilities. Learning outcomes can be measured using the right instruments. This article discusses the analysis of the test results in the form of multiple choice questions (dichotomy scores). This research is a descriptive quantitative research that describes the results of quantitative data analysis. The data used in the form of secondary data, namely the National Examination (UN) questions for SMP mathematics subjects with a sample of 1000 respondents and 7 items of UN questions selected randomly. Data analysis was performed using the R program with the Rasch model selected with one logistic parameter and calculating the MLE. From the results of the analysis, it was obtained that the item difficulty index on the easy and medium criteria, from the sample taken, there were 109 response patterns for the test taker's answers. Then analyzed one of the response patterns, namely (0 0 1 1 0 0 0) and it was concluded that the test takers with this pattern had a chance to answer correctly below 50% so that they might not have mastered the concept, guessed or cheated on the answer. Suggestions for teachers and lecturers to carry out a similar analysis or a more detailed analysis to see the characteristics of the items made and the abilities of their students.