multiple regression

[parhamed]

---

Members don't see this ad.

Hello friends,
I have 2 IV, one DV and 1 moderator and want to know influence of moderator on the relationship between IV and DV, i used SPSS: first i entered IV and moderator and in the second step i entered interaction between them. i used stepwise method? i want to be sure this process is correct or not? actually we can use stepwise method?

 

[Ollie123]

I wouldn't use stepwise for something like this - even if the main effects aren't significant, you still want them in the model if you are testing the interaction. The inverse isn't true though - when reporting the final model its acceptable (and generally preferred depending on your design) to back the interaction term out if it isn't significant for easier interpretation of the main effects.

 

[Sunnysideup03]

Is one of your IV's a covariate? Or are you creating a 3-way interaction?
If one of the IV is a covariate I would put that in the first step, then the second IV (that you created your interaction term with), then the moderator, then the interaction term. It's a lot of steps but I just have always done it like that - I don't think your results will change even if you put them all together, I just like things very organized in my output - I get confused easily.

If you're creating a 3-way interaction then you will need to enter the IVs and moderator, then 2-way interactions of the IVs and DVs (all possible combinations), then the final step would include the 3-way interaction.

Does that make sense? Just PM me if it needs further clarification - stats can be one of the most complicated parts of research if you ask me.

Good luck! hope you have a significant interaction 🙂