How do I run multiple regression on two independent and one dependent variable using SPSS?
So, I have seven items in construct 1 which is an independent variable, four-item in another independent variable which is a construct 2, and then six-item in construct 3 which is a dependant variable.
I am not being able to run the 2 independent constructs with the 1 dependant construct and two control variable age and gender.
I need to run a multiple regression. With 300 plus respondents.
Any pointers? Should I use a summation of the responses?
Please look in package manual under multiviate linear model.
Or google same. D. Booth PS consulting a social science research methods text would probably help too. D. Booth
Hello Quazi,
Professor Booth is pointing you in the right direction: explanations of how to set up GLM models.
However, the first big question you'll need to resolve before proceeding is to decide whether it makes more sense (conceptually, measurement-wise, and with respect to your research question/s) to treat each of these three variables as a single score or as a set of related scores (representing the responses to the constituent items).
I'd suggest you're likely better off with the former approach (single scores) in which case the model should resolve to a pretty straightforward multiple regression. How to generate that single score becomes the second question before hammering on the data. Your options would be: (a) some simple linear combination (summing or averaging the respective item scores, for example); (b) deriving factor score estimates (either from factor analysis or SEM approach); or (c) using IRT to derive a scale score for each respective variable.
Good luck with your work.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS