Loading…
Comparison of ability parameter estimation in dichotomous and polytomous item response models (based on national exam scores in mathematics)
This study aims to compare the estimation of test takers’ ability in the polytomous and dichotomous assessment models. The data used in this study are secondary data on the results of the 2014/2015 national exam in Mathematics as many as 40 items, with a sample of 3000 test participants taken random...
Saved in:
Main Authors: | , , |
---|---|
Format: | Conference Proceeding |
Language: | English |
Subjects: | |
Online Access: | Get full text |
Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
Summary: | This study aims to compare the estimation of test takers’ ability in the polytomous and dichotomous assessment models. The data used in this study are secondary data on the results of the 2014/2015 national exam in Mathematics as many as 40 items, with a sample of 3000 test participants taken randomly. Analysis for dichotomous scores using BILOG software and for polytomous scores using PARSCALE software. Some of the points analyzed were item characteristics and test takers’ abilities. Dichotomous scoring is done by giving a score of 1 for correct answers, and a score of 0 for wrong answers. While for the polytomous score is obtained by scoring with the PCM model, where several items are grouped based on the same indicator so that 18 items are obtained, with calculations carried out per item in each indicator, if answered correctly gets a score of 1, and if answered incorrectly gets a score of 0. The results of the analysis of the dichotomous score model using BILOG software showed that the average ability of test takers was 3.6×10−6, the minimum ability was -3.2254 and the maximum ability was 1.4941, with the standard deviation for the test taker’s ability was 1.000. The results of the analysis of the polytomous score model using PARSCALE software, the results showed that the average ability of test takers was −5.033×10−6, the minimum ability was -1.4971 and the maximum ability was 1.7994 with a standard deviation of test takers’ ability of 1.000. Based on the results of the analysis, it can be concluded that both polytomous and dichotomous estimation of ability is included in the moderate category and dichotomous calculation is better than polytomous. Suggestions for further research, the calculation of the estimated ability of test takers can be done using different software. |
---|---|
ISSN: | 0094-243X 1551-7616 |
DOI: | 10.1063/5.0194639 |