Implementation of Item Response Theory for Analysis of Test Items Quality and Students’ Ability in Chemistry

Abstract

This first aim of this study is to describe the quality of chemistry test item made by teacher. The test was developed for 11 th grade students’ science class in the first semester on academic year 2015/2016. The second aim of this study is to describe the characteristic of measurement’s result for students’ ability in chemistry. This is descriptive research design with the 101 student’s responses patterns from multiple choice test device with 5 answer alternatives. The responses patterns were collected by documentation technique and analyzed quantitatively using Item Response Theory software such as BILOG MG V3.0 with 1-PL, 2-PL, and 3-PL models. The differences of students’ ability in chemistry in model 1-PL, 2-PL, dan 3-PL were analyzed using One-Way Anova Repeated Measure. The result showed that the mean of item difficulties level (b), item differentiate (a), and pseudo-guessing (c) are good. The measurement tools arranged by teacher were suitable for students who have the ability from -1.0 to +1.7. The maximum score of item information function is 68.83 (SEM =0.121) with ability in 0.2 logit. The highest ability’s estimation score was showed by Model 2-PL. The mean of students’ ability for 11 th grade students is -0.0185 logit and consider as moderate category.