因子分析元分析的三阶段整合模型

因子分析元分析(meta-analysis of factor analyses, MAFA)整合了因子分析和元分析两种方法学传统。开展MAFA方法学研究可结束过去几十年MAFA研究缺乏根据的状况, 服务心理学等领域内的知识生产。三阶段整合模型认为MAFA包括数据准备、数据合成和数据分析三个主要阶段。模型及数据合成技术的有效性将用心理测量学方法, 包括蒙特卡洛模拟研究进行检验。研究结果将阐明正确运用MAFA的一般规则、操作步骤和注意事项, 进而增进我们对MAFA的认识, 丰富应用心理等领域的方法学体系, 为MAFA方法的推广应用奠定基础。

A Three-stage Integrative Model of Meta-analysis of Factor Analyses

Meta-analysis of factor analysis (MAFA), also called meta-analytic factor analysis, integrated two methodological traditions namely meta analysis and factor analysis. Research on MAFA could explore some methodolical issues and accelerate the production process of psychological knowledge which will end the theoretical-principle-absence status of the meta analytic factor analysis had existed for several decades. To represent the process of MAFA, the present project put forth a three-stage integrative model under which some scientific questions will be tested using psychometrical methods including Monte Carlo simulation study. The validity of the three-stage integrative model and that of its data-synthesis methods (i.e. the co-occurrence matrix of salient factor loadings, the aggregate correlation matrix, and the pairwise rotation of results to congruence) were the key problems in this project. The project also planned to explore the general rules theoretically and steps of how to conducut MAFA study whose results would help the research community to use MAFA correctly. The possible scientific contribution of the present project is that it will add to the store of meta-analytic and factor analytic traditions, enrich the content of psychological methodology, and establish a solid base to popularize the MAFA.