I have conducted a moderation analysis as follows:

x (continuous independent variable, centered) m (continuous moderator, centered)
y (dependent variable)

By entering x, m, and the product of x and m (x.m, interaction factor) into simultaneous regression, I get an overall significant model, with significant regression coefficients for x (beta x=0.273), for m (beta m=0.215) and for x.m (beta x.m=-0.034), all p<0.05. I understand that this means that m is a significant moderator of the relationship between x and y.

However, when I try to plot simple slopes at high (+1SD) and low (-1SD) levels of m, both slopes are nonsignificant (unstandardized slope 0.029, t=0.314 and p=0.89 for high m, and unstandardized slope 0.1349, t=0.902 and p=0.37 for low m). I believe this means neither slope is different from zero.

My question is, does this invalidate my interaction? Should I therefore not consider m a significant moderator of y on x? If the interaction is still valid, how do I report the results of the simple slope analysis?

  • 1
    $\begingroup$ You say in the first full paragraph "all p>.05". Did you mean that all p values were less than .05 (i.e. "all p < .05")? You also refer to both x and y as "independent variables", is y a dependent variable? Finally, could you say whether the standard deviation of x is smaller than 1? $\endgroup$ – wools Jun 9 '15 at 3:24
  • $\begingroup$ Yes, p-values in the first paragraph less than 0.05, and y is a dependent variable. I have edited the post to correct those mistakes. Standard deviation of x is 3.8. $\endgroup$ – B B Jinx Jun 9 '15 at 11:10
  • $\begingroup$ There are many questions somewhat like this, have a loook at stats.stackexchange.com/questions/11009/… and the list stats.stackexchange.com/search?q=interaction+significant+but+ $\endgroup$ – kjetil b halvorsen Jun 9 '15 at 14:17
  • $\begingroup$ Please register & merge your accounts. Then you will be able to edit & comment on your own question. You can find out how to do this in the My Account section of our help center. $\endgroup$ – gung - Reinstate Monica Jun 9 '15 at 14:25
  • $\begingroup$ To kjetil b halvorsen, thanks for your response; I did do an extensive search before posting my question and none of the other entries seemed to apply to my situation. They deal with scenarios in which the interaction term is not significant but simple slopes are, or the main effects are nonsignificant. In my situation, the interaction term is significant but the simple slopes are not. My question is, does that invalidate the significant interaction? Thanks again. $\endgroup$ – B B Jinx Jun 9 '15 at 15:02