I have two continuous variables; a pretest and a post test results. They seem to have significance in paired samples; p=.000. However I wonder what caused the post test to be better than the pre test and I suspect that another continuous variable predicted the increase in post test. I restructured my data to long format in spss and now have a single variable for 2 tests and a categorical index variable (1-2). I am trying to build up a regression equation in which another continuous variable is a covariate predictor. I am not sure what to do exactly, I need some kind of repeated measures regression in SPSS if possible. Any suggestions are welcome. Thanks a lot.
Emrah, you did not specify what is the main problem of research. If the goal is to test the effect of changes in a continuous variable (two dependent series of results) and two independent groups (categorical variable), you should use the analysis of repeated measurements or analysis of covariance. If you suspect that some other variable significant influence on these changes, it should be included as covariates in the analysis. In case you have had some experimental treatment, its impact on changes in continuous variables are registered through testing differences between groups (experimental and control) at the end of treatment. The effect of changes in the group you can test with t test for paired samples.
I hope this explanation will help you.
Supposedly, in an experiment, the "treatment" or intervention is hypothesized to cause the change in the experimental group. The control group is included to find out if there could be factors other than the treatment.
I would advise that you refer to the research design instead of guessing the cause of the change.
It seems difficult to help you with the information you gave. Can you détail the experiment? The other factors that could influence the changes? What do you measure? Observe? Could you provide a sample of you dataset?
Thanks a lot for answers Zeljko, Eddie and Alex. I solved the design with GEE Logistic regression by adding subject and within subject variables.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS