Moderation analysis using the Process tool in SPSS.

#1
Dear All,

I was wondering if any one could help me with the interpretation of the following output. As you can see in the first picture, the interactionterm is non-significant.. What is confusing to me is that, as you can see in the second picture, the predictor does interact with the moderator at 1 standarddeviation above the mean of this variabel. How can that be? Does this mean that there IS an interaction? Or is there non? 1592928476190.png 1592928708235.png
 

katxt

Active Member
#2
I'm not an SPSS user and but in general, if an interaction (or any factor) is shown to be insignificant, then that is the end of it. You cannot split up the various components of it using post hoc tests. In effect, the final three p values give you three chances at a p value less than 0.05 so your 95% protection against a false positive has been eroded. You could use Bonferroni and set the critical value for significance over these three p values at 0.05/3, or say 0.02. In this case, the p = 0.0428 is not significant. There may well be an effect but this is not convincing evidence of it.
 
#3
I'm not an SPSS user and but in general, if an interaction (or any factor) is shown to be insignificant, then that is the end of it. You cannot split up the various components of it using post hoc tests. In effect, the final three p values give you three chances at a p value less than 0.05 so your 95% protection against a false positive has been eroded. You could use Bonferroni and set the critical value for significance over these three p values at 0.05/3, or say 0.02. In this case, the p = 0.0428 is not significant. There may well be an effect but this is not convincing evidence of it.
Thank you for your reply