" V: W8 x" D3 X% k' V1. 以后请发到圈子的帖子。我很少看涂鸦的。 , V# T% X+ |* G$ q2. 不可以，理论上应该做 full model testing，把所有的关系一同测验。 & x. h% M' b5 J* x3. 我的理解是中心化已经是一个标准了。 W, e. u$ @7 t: n你可以把Hayes的整个reference 给我，让我看看吗5 G) j' _2 i" P A1 b& M: z
; a; L f& }3 o+ C
Hayes：[size=11.818181991577148px]The explanation that seems to have resulted in the most misunderstanding is that X and M are likely to be highly correlated with XM and this will produce estimation problems caused by multicollinearity and result in poor or “strange” estimates of regression coefficients, large standard errors, and reduced power of the statistical test of the interaction. This is, in large part, simply a myth. As described later, there are some reasons that centering can be a beneficial thing to do, which is why it has been recommended by some. / T: ~( y! G8 E5 |% {1 M# nHowever, it is incorrect to claim that it is necessary, that a failure to do so will lead one to incorrect inferences about moderation, or that the resulting regression coefficients are somehow strange or uninterpretable. & G+ V& f- e4 P+ b/ R6 c , t0 n; z1 R7 O4 C作者: Kenneth 时间: 2014-9-29 21:58
秋水长空， 2 @7 M5 M+ G9 C& t0 K$ D# @谢谢你的 reference。) }9 R8 [4 X* I- O* J# I0 s
但是，我们做实证研究是讲求证据的。如果 Hayes 是单单说 “it is incorrect to claim that it is necessary, that a failure to do so will lead one to incorrect inferences about moderation, or that the resulting regression coefficients are somehow strange or uninterpretable.” 而没有提出证据的话，我觉得有点不负责任。他有解释吗？