基于多元回归的调节效应分析  

Moderation Effect Analyses Based on Multiple Linear Regression

作　　者：方杰[1] 温忠麟[2] 梁东梅[2] 李霓霓[3] 

机构地区：[1]广东财经大学人文与传播学院,广州510320 [2]华南师范大学心理学院/心理应用研究中心,广州510631 [3]广州美术学院,广州510260

出　　处：《心理科学》2015年第3期715-720,共6页Journal of Psychological Science

基　　金：国家自然科学基金项目(31271116);教育部人文社会科学研究青年基金项目(14YJC190003);广东省哲学社会科学"十二五"规划项目(GD13CXL01)的资助

摘　　要：在心理学和其他社科研究领域,大量实证研究建立调节模型,以分析自变量对因变量关系的影响机制,但在基于多元回归的调节效应分析实践中仍存在不足。我们回顾了均值中心化在基于多元回归的调节效应分析中的作用,均值中心化不影响乘积项(即调节效应)的检验,仅对一阶项(即主效应)的检验有影响。讨论了简单斜率的检验方法,建议在调节变量为连续变量时,使用Johnson-Neyman法进行简单斜率检验;在调节变量为类别变量或研究者对某个调节变量值感兴趣时,使用选点法。并用一个实际例子演示如何进行调节效应分析。随后展望了调节效应检验的拓展方向。Moderation indicates that the strength and/or direction of the relation between an independent variable and a dependent variable is affected by a third variable called the moderator. Moderation models are frequently used in the research of psychology and other social science disciplines, still some issues need to be clarified. The purpose of the present study is to clarify two issues in the moderation effect analyses. One is the role of mean-centering for original variables in moderation modeling; the other is the advantages and disadvantages of two existing methods for testing the simple slope. Firstly, the product term in moderated regression might be collinear with its constituent parts, making it difficult to detect moderation effects. Some researchers presumed that mean-centering could reduce coUinearity and improve the precision of estimates from collinear data, but this is not true. After reviewing the role of mean-centering in moderated multiple regression, we emphasize that mean-centering does not change the coefficient of the product term （moderation term） of the regression, but changes the coefficients of the first-order terms （main effect terms） and improves the interpretability of results. Secondly, when a moderation effect is found, the moderation effect needs to be further probed to fully explicate the relationship among the three variables. The most common method for probing the moderation is to test the simple slopes. We discuss the merits and demerits of two methods for testing the simple slopes： the Pick-a-point method and the Johnson-Neyman＇s method. The Pick-a-point method is to test simple slopes at several specific levels of the predictors and to report whether they are significant or not, whereas the Johnson-Neyman＇s method is to test simple slopes in the whole range of the predictor and to report the regions in which the simple slopes are significant. We suggest that the Johnson- Neyman＇s method be adopted to test the simple slopes when the moderator is a continuous varia

关 键 词：调节效应 多元线性回归 均值中心化 选点法 Johnson-Neyman法 

分 类 号：O212.1[理学—概率论与数理统计]