Estimation of IRT Graded Response Models: Limited Versus Full Information Methods
Estimation of IRT Graded Response Models: Limited Versus Full Information Methods
复制标题
DOI：
10.1037/a0015825
复制
发表时间：
2009-09-01
影响因子：
7
通讯作者：
Maydeu-Olivares, Alberto
中科院分区：
文献类型：
--
作者：
Forero, Carlos G.;Maydeu-Olivares, Alberto
讨论了F. Samejima的分级反应模型在324种条件下进行了检查。当分类项目因子分析（CIFA）的第三阶段使用未加权最小二乘法时，将完全信息最大似然（FIML）与3阶段估计进行了比较。CIFA在估计多维模型时要快得多，特别是具有相关维度的模型。总的来说，CIFA产生的参数估计值稍微更准确，FIML产生的标准误差稍微更准确。然而，在大多数情况下，方法之间的差异可以忽略不计。FIML是小样本量（200个观测值）的最佳选择。CIFA是最好的选举在较大的样本（计算的理由）。这两种方法在许多情况下都失败了，其中大多数涉及200个观察结果，每个维度的指标很少，高度倾斜的项目或低因子负荷。在应用中应避免这些情况。
The performance of parameter estimates and standard errors in estimating F. Samejima's graded response model was examined across 324 conditions. Full information maximum likelihood (FIML) was compared with a 3-stage estimator for categorical item factor analysis (CIFA) when the unweighted least squares method was used in CIFA's third stage. CIFA is much faster in estimating multidimensional models, particularly with correlated dimensions. Overall, CIFA yields slightly more accurate parameter estimates, and FIML yields slightly more accurate standard errors. Yet, across most conditions, differences between methods are negligible. FIML is the best election in small sample sizes (200 observations). CIFA is the best election in larger samples (on computational grounds). Both methods failed in a number of conditions, most of which involved 200 observations, few indicators per dimension, highly skewed items, or low factor loadings. These conditions are to be avoided in applications.