第八讲中介效应及其检验msci2019.pptxVIP

第八讲 中介效应及其检验吴建祖 教授兰州大学管理学院中介变量吴建祖 博士中介变量（mediator）是自变量对因变量发生影响的中介。中介效应的研究主要是为了打开两个变量之间关系的『黑盒子』，解释变量之间关系的内部机理。McKinnon DP. 2008. Introduction to statistical mediation analysis. Taylor and Francis Group: New York.论文样例吴建祖 博士中介效应路径模型和回归方程吴建祖 博士检验中介效应的回归方程吴建祖 博士中介效应Total effect = Direct effect + Indirect effectTotal effect, cDirect effect, c’Indirect effect (Mediated effect), a*b总效应=直接效应+间接效应（中介效应）总效应, c直接效应, c’间接效应（中介效应）, a*b吴建祖 博士证明：总效应=直接效应+间接效应（中介效应）吴建祖 博士完全中介 vs. 部分中介步骤条件完全中介部分中介(1)以X预测Y预测效果存在预测效果存在(2)以X预测M预测效果存在预测效果存在(3)同时以X与M预测YM的预测效果仍存在；X的预测效果消失X与M的预测效果仍存在XMYXMY完全中介(Complete mediation)部分中介(partial mediation)吴建祖 博士中介效应的检验方法吴建祖 博士Barron Kenny模型Sobel检验BootstrappingBarron Kenny模型中介效果的检验下属的工作满意度主管的不当对待领导下属的公平知觉(X, 自变量)(M,中介变量)(Y, 因变量) 三部曲(1)以X去预测Y(2)以X去预测M(3)同时以X与M去预测YBased on Baron Kenny(1986); Also see Shaver(2005)吴建祖 博士Barron and Kenny, 1986吴建祖 博士Step 1: Relation of X to YMEDIATORMINDEPENDENT VARIABLEDEPENDENT VARIABLEcXYThe independent variable is related to the dependent variable:Y = i1 + cX + e1吴建祖 博士Step 2: Relation of X to MMEDIATORMaINDEPENDENT VARIABLEDEPENDENT VARIABLEXY2. The independent variable is related to the potential mediator:M = i2 + aX + e2吴建祖 博士Step 3: Relation of X and M to YMEDIATORMabINDEPENDENT VARIABLEDEPENDENT VARIABLEc’XY3. The mediator is related to the dependent variable controlling for exposure to the independent variable:Y = i3+ c’X + bM + e3吴建祖 博士Sobel TestSobel Test吴建祖 博士A Example of Sobel Test/sobel/sobel.htm/sobel/sobel.htm吴建祖 博士Sobel Test吴建祖 博士The Sobel test works well only in large samples.We recommend using this test only if the user has no access to raw data.If you have the raw data, bootstrapping offers a much better alternative that imposes no distributional assumptions.Preacher K, Hayes A. 2004. SPSS and SAS procedures for estimating indirect effects in simple mediation models. Behavior Research Methods, Instruments, Computers 36(4): 717-731.BootstrappingDistribution of the ProductThe mediated effect is the product of