Hello everyone!
Is it possible on SPSS to test a moderation model with:
- a dummy independent (0 vs 1)
- a continuous moderator (measured on a 7 points Likert scale)
- a continuous independent variable (measured on a 7 point likert scale)?
I am really struggling with this and I hope some of you can help me.
Thanks a lot!
Hello Margherita Carotti
See the link below. It has a good explanation for you.
https://statistics.laerd.com/spss-tutorials/dichotomous-moderator-analysis-using-spss-statistics.php
Of course, this link is also good
https://www.spss-tutorials.com/spss-regression-with-moderation-interaction-effect/
Hello Margherita Carotti. I think you will find it more convenient to use the UNIANOVA command rather than the REGRESSION command, as it has nice features you can use (e.g., EMMEANS subcommand) to generate output that helps you interpret the interaction. And by including /PRINT=PARAMETER, you'll get the same table of coefficients you see when using REGRESSION.
I have attached a syntax file that shows an example using one of the datasets that comes with SPSS. Just change the path to where you have the sample datasets stored on the GET FILE line, and then Run > All.
It should not be too difficult to modify the code to make it suitable for your own variables.
HTH.
Dear Margherita,
simply create new variable in the dataset that is a result of a simple multiplication of the dummy with the moderator. Then add this product variable together with the dummy, moderator, and other control variables.
Some folks like it in a hierarchical (better: stepwise) procedure with step 1 including the controls, step 2 including all main effect variables and step 3 including the product variable.
How to interpret the results is IMHO best explained in the Book by Aiken & West (1991):
Aiken, L. S., & West, S. G. (1991). Multiple regression. Newbury Park, CA: Sage.
All the best
Holger
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS