论文部分内容阅读
元分析是根据现有研究对感兴趣的主题得出比较准确和有代表性结论的一种重要方法,在心理、教育、管理、医学等社会科学研究中得到广泛应用。信度是衡量测验质量的重要指标,用合成信度能比较准确的估计测验信度。未见有文献提供合成信度元分析方法。本研究在比较对参数进行元分析的三种模型优劣的基础上,在变化系数模型下推出合成信度元分析点估计及区间估计的方法;以区间覆盖率为衡量指标,模拟研究表明本研究提出的合成信度元分析区间估计的方法得当;举例说明如何对单维测验的合成信度进行元分析。
Meta-analysis is an important method to derive more accurate and representative conclusions based on the existing research. It is widely used in the social sciences such as psychology, education, management and medicine. Reliability is a measure of the quality of an important indicator of the test, the reliability of the synthesis can be more accurate estimation test reliability. No literature has provided synthetic reliability analysis method. Based on the comparison of the merits and demerits of the three models of the meta-analysis of the parameters, this study put forward the method of the point analysis and the interval estimation of the composite reliability analysis under the variation coefficient model. Taking the interval coverage as the measurement index, The method proposed in this paper is suitable for the interval estimation of the composite reliability analysis. The example shows how to perform the meta-analysis of the composite reliability of the one-dimensional test.