Astudy on diabetes has two groups, one control and other case, the question is does it make any sense to do correlation by combining the two groups or is it better to do correlation seperatly group wise? Is there any condition that warrents the combination of two groups while doing correlation?
Given that you are using SPSS, you can also make a scatter-plot and set markers by your group variable. Then you can double-click on the graph to open the chart editor, and add fit lines by sub-group, or something to that effect. See example 3 at the link below to get an idea about how to do it. HTH.
http://www.ats.ucla.edu/stat/spss/library/hetreg.htm
The basic question is whether the correlation would be the same or different in the two groups. If the relationship differs across the two groups, then you should not combine them.
You can test this by using dummies variables and an interaction effect in multiple regression.
Given that you are using SPSS, you can also make a scatter-plot and set markers by your group variable. Then you can double-click on the graph to open the chart editor, and add fit lines by sub-group, or something to that effect. See example 3 at the link below to get an idea about how to do it. HTH.
http://www.ats.ucla.edu/stat/spss/library/hetreg.htm
for me, I would not combine the 2 group together. it also depend on your targeted variables.
- Badges
- Science topic
- Similar topics
- Mathematical Sciences
- Statistics
- Statistical Software
- SPSS