Dimitrov Dimiter M
Graduate School of Education, MNN4B3, 4400 University Dr., George Mason University, Fairfax, VA 22030, USA.
J Appl Meas. 2003;4(3):222-33.
This article provides formulas for expected true-score measures and reliability of binary items as a function of their Rasch difficulty when the trait (ability) distribution is normal or logistic. The proposed formulas have theoretical value and can be useful in test development, score analysis, and simulation studies. Once the items are calibrated with the dichotomous Rasch model, one can estimate (without further data collection) the expected values for true-score measures (e.g., domain score, true score variance, and error variance for the number-right score) and reliability for both norm-referenced and criterion-referenced interpretations. Thus, given a bank of Rasch calibrated items, one can develop a test with desirable values of population true-score measures and reliability or compare such measures for subsets of items that are grouped by substantive characteristics (e.g., content areas or strands of learning outcomes). An illustrative example for using the proposed formulas is also provided.
本文给出了在特质（能力）分布为正态或逻辑分布时，作为二分项目Rasch难度函数的期望真分数测量值和信度的公式。所提出的公式具有理论价值，可用于测试开发、分数分析和模拟研究。一旦用二分Rasch模型对项目进行校准，就可以（无需进一步收集数据）估计真分数测量值的期望值（例如，领域分数、真分数方差和答对分数的误差方差）以及常模参照和标准参照解释的信度。因此，给定一组经Rasch校准的项目，就可以开发出具有理想总体真分数测量值和信度的测试，或者比较按实质特征（如内容领域或学习成果的维度）分组的项目子集的此类测量值。还提供了一个使用所提出公式的示例。
