Obtaining a common scale for item response theory item parameters using separate versus concurrent estimation in the common-item equating design
Obtaining a common scale for item response theory item parameters using separate versus concurrent estimation in the common-item equating design
复制标题
DOI：
10.1177/0146621602026001001
复制
发表时间：
2002-03-01
影响因子：
1.2
通讯作者：
Béguin, AA
中科院分区：
文献类型：
--
作者：
Hanson, BA;Béguin, AA
项目反应理论项目参数可以估计使用数据从一个共同的项目等值设计，无论是单独为每个表格或同时跨表格。本文报道了一个模拟研究的结果，单独与并发项目参数估计。使用来自60个二分项目的测试的模拟数据，考虑了四个因素：（a）估计程序（MULTIPLE与BILOG-MG），（B）每个表格的样本量（3，000与1，000），（c）共同项目的数量（20与10），以及（d）采用两种表格的等效组与非等效组（无平均差异与平均差异I SD）。此外，在不同的估计条件下，采用了四种项目参数标度方法：两种项目特征曲线法（Stocking-Lord和Haebara）和两种矩法（Mean/Mean和Mean/Sigma）。同时估计一般会导致较低的误差比单独估计，虽然不是普遍如此。结果表明，一个因素占较低的误差时，使用并发估计可能是参数估计的共同项目参数是基于较大的样本。有人认为，这项研究的结果，连同其他研究在这个问题上，是不足以建议完全避免单独的估计，有利于并发估计。
Item response theory item parameters can be estimated using data from a common-item equating design either separately for each form or concurrently across forms. This paper reports the results of a simulation study of separate versus concurrent item parameter estimation. Using simulated data from a test with 60 dichotomous items, four factors were considered: (a) estimation program (MULTILOG versus BILOG-MG), (b) sample size per form (3,000 versus 1,000), (c) number of common items (20 versus 10), and (d) equivalent versus nonequivalent groups taking the two forms (no mean difference versus a mean difference of I SD). In addition, four methods of item parameter scaling were used in the separate estimation condition: two item characteristic curve methods (Stocking-Lord and Haebara) and two moment methods (Mean/Mean and Mean/Sigma). Concurrent estimation generally resulted in lower error than separate estimation, although not universally so. The results suggest that one factor accounting for the lower error when using concurrent estimation may be that the parameter estimates for the common item parameters are based on larger samples. It is argued that the results of this study, together with other research on this topic, are not sufficient to recommend completely avoiding separate estimation in favor of concurrent estimation.