thank you! but the interaction effect you mentioned should not be equal to the mediating effect! in fact, moderating effect is one type of interaction effect, which suggects how much effect the independent variable has on dependent varialbe relys on the third variable (the moderating variable). It should be mentioned that moderating effect has its direction, however, the pure interaction effect has not.
Further, we should differ the mediating effect from moderating effect. Note that a mediational model is a causal model. For example, independent variable (A) has an impact on the dependent variable (C) through the mediating role of B. In a word, the mediator (B) can deliver the impact that independent variable (A) offered to the dependent varialbe (C). The path should be "A---->B----->C". however,a moderator variable M is a variable that alters the strength of the causal relationship. So for instance, psychotherapy may reduce depression more for men than for women, and so we would say that gender (M) moderates the causal effect of psychotherapy (X) on depression (Y). The path should be "X*M--->Y"
anyway, thank you for you reply!
作者: mischina 时间: 2011-7-2 22:15
刚看调节变量,现学现卖,一起探讨。 如果自变量X,因变量Y。调节变量为M,那么所谓调节方程应为 Y = aX + bM + cXM + e 重新写成 Y = bM + ( a + cM ) X + e 对于固定的M ,这是Y对X 的直线回归。Y与X 的关系由回归系数a + cM 来刻画,它是M 的线性函数, c衡量了调节效应(moderating effect)的大小。故也即C值也就是你所说的调节变量是增还是减。 另外还要注意调节效应和中介效应的区别,我认为你所说的应是正确。 我推荐一篇文篇给你看 调节效应与中介效应的比较和应用(温忠麟1, 2 侯杰泰2 张 雷) 另外,前面的贴子里可能也有相关的题,如 http://bbs.pinggu.org/thread-358719-1-1.html你可以参考 (个人理解:里面所说的离均差应是中心化,也就原观察值减去观察值的均值) 不外,上面所说的外文文献我都没有找到,如果你找到了可以发给我,谢了!lrfdnv@163.com 我也是刚刚学,说错地方多多见谅。
thank you! but the interaction effect you mentioned should not be equal to the mediating effect! in fact, moderating effect is one type of interaction effect, which suggects how much effect the independent variable has on dependent varialbe relys on the third variable (the moderating variable). It should be mentioned that moderating effect has its direction, however, the pure interaction effect has not.
Further, we should differ the mediating effect from moderating effect. Note that a mediational model is a causal model. For example, independent variable (A) has an impact on the dependent variable (C) through the mediating role of B. In a word, the mediator (B) can deliver the impact that independent variable (A) offered to the dependent varialbe (C). The path should be "A---->B----->C". however,a moderator variable M is a variable that alters the strength of the causal relationship. So for instance, psychotherapy may reduce depression more for men than for women, and so we would say that gender (M) moderates the causal effect of psychotherapy (X) on depression (Y). The path should be "X*M--->Y"
anyway, thank you for you reply!
作者: mischina 时间: 2011-7-2 23:20
刚看调节变量,现学现卖,一起探讨。 如果自变量X,因变量Y。调节变量为M,那么所谓调节方程应为 Y = aX + bM + cXM + e 重新写成 Y = bM + ( a + cM ) X + e 对于固定的M ,这是Y对X 的直线回归。Y与X 的关系由回归系数a + cM 来刻画,它是M 的线性函数, c衡量了调节效应(moderating effect)的大小。故也即C值也就是你所说的调节变量是增还是减。 另外还要注意调节效应和中介效应的区别,我认为你所说的应是正确。 我推荐一篇文篇给你看 调节效应与中介效应的比较和应用(温忠麟1, 2 侯杰泰2 张 雷) 另外,前面的贴子里可能也有相关的题,如 http://bbs.pinggu.org/thread-358719-1-1.html你可以参考 (个人理解:里面所说的离均差应是中心化,也就原观察值减去观察值的均值) 不外,上面所说的外文文献我都没有找到,如果你找到了可以发给我,谢了!lrfdnv@163.com 我也是刚刚学,说错地方多多见谅。