Item Analysis of English Summative Test Based On Item Response Theory (IRT)

This study aims at finding out the indices of English summative test as well as
describing the test quality based on the 3-PLM of IRT (Item Response Theory). A
descriptive-quantitative design is applied by using Rstudio program to analyze the
data. The findings show that there are three indices in evaluating English test
based on the 3-PLM of IRT, they were item difficulty (b -parameter), item
discrimination (a -parameter) and pseudo guessing (c -parameter). As much as 95
% (38 items) are in the good category of difficulty index and 5% (2 items) are in
the poor category/very easy items. Then, as much as 30 % (12 items are in the
good category and 70 % (28 items) in the poor category of the discrimination
power. While referring to the guessing parameter, as much as 90 % (36 items) are
in the good category and 10 % (4 items) are in the poor category.