- 期刊
A Comparison of Methods of Estimating Subscale Scores for Mixed-Format Tests
摘要
Purpose: The purpose of this study is to compare methods of estimating subscale scores using mixed-format tests. Design: A simulation study is conducted with four factors (the correlation between subscales, the number of items per subscale, the ratio of multiple choice/constructed response [MC/ CR] items, and the number of examinees) using score reliability, 95% confidence/credibility interval (95CI), the percentage-coverage of a nominal 95% confidence/credibility interval (95PC), absolute bias, standard error of estimation, and root mean square error (RMSE) as evaluation criteria. Findings: From the results of this study, using collateral test information to estimate objective scores did increase the reliability and reduce the width of 95CI and the RMSE compared to the proportion-correct method. Within the methods studied, the Wainer and Shin methods yielded subscale scores that had the highest reliability and lowest RMSE. The Yen method had the narrowest 95CI and relatively accurate 95PC. Practical implications: Subscale scores are the most common type of diagnostic information included in student score reports. Because the number of items contributing to subscales is usually small, several procedures currently exist to improve the reliability of subscale scores. This study extended these procedures to mixed-format tests and compared their performance. The finding can provide practical information for the practitioners that would like to use these methods.