结构方程建模中的题目打包策略

结构方程建模中题目打包法的优缺点包括：指标数据质量变好、模型拟合程度提高; 估计偏差不大, 可校正; 估计稳定, 但降低了敏感性与可证伪性。打包法的前提条件是单维、同质, 适合结构模型分析, 不适合测量模型分析。对于单维测验, 给出了一个打包流程。对于通常的多个子量表(多维结构)测验, 推荐在子量表内打包, 每个子量表打包成1个指标或者3个指标, 用于结构方程建模。

Item Parceling Strategies in Structural Equation Modeling

Item parceling is a technique using in structural equation modeling (SEM). Parceling can improve the quality of indicators and model fit. Bias that due to parceling was often neglectable and can be corrected. Parceling greatly enhances model parsimony, but it greatly reduces falsifiability of the tested model. It could be summarized that the prerequisites of parceling are unidimension and homogeneity, and the applicability of parceling is the analysis of structural models, rather than measurement models. Parcel-building algorithms and the number of parcels were discussed and recommended. A procedure for item parceling was proposed when the scale was unidimensional. If the scale was multidimensional, internal-consistency approach was recommended such that the items of the same dimension are parceled to one or three indicators for structural equation modeling.