Hello,
I am currently working on my Bachelor thesis and would like to find out how I should compute this on SPSS and which test I should use to find out how each group affects my dependent variable.
I want to create three different groups:
Group 1: High Score on x and low score on y and their combined effect on the outcome variable
Group 2: High score on x and high score on y and their combined effect on the outcome variable
Group3: Low Score on x and low score on y and their combined effect on the outcome variable
Group 4: Low Score on x and high score on y and their combined effect on the outcome variable
Each help is welcomed!
Thanks
Hello Lisa,
I see a couple of options.
1. Ordinary, 2 x 2 factorial anova. The question of whether specific combinations of x and y levels are important for accounting for differences on your dependent variable is addressed by the test of interaction of the two factors. This presumes that "level" (e.g., high or low) means that x and y are already dichotomized, or that you will convert some continuous variable into two levels (not recommended).
2. Multiple linear regression. The independent variables would be value on x, value on y, and an interaction of x and y (this can be formed by multiplying x by y for each case to create a product term--some folks like to "center" these scores, by multiplying (x - mean of x) by (y - mean of y) when forming the product term). This will work regardless of whether x and y are dichotomous or continuous. Again, the point that's important to your query is whether the interaction term is significant.
Good luck with your work.
Prof Morse is correct but I advise moderation (interaction) plots as well.
In addition you might find the attached information useful. I would suggest adding SPSS and interaction plot to your google query. Don't forget that this is all from factorial experimental designs So you could look at these as well. Best wishes, David Booth
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS