Hello,
As you know, in logistic regression models, outputs of factor analysis (summated scales and factor scores) can be used as independent variables?
In a thesis study, researcher suggests using summated scales as independent variables. Which one do you suggest?
Thanks on advance
Well it depends on the research question, and the statistical methods employed to say which is better. However researchers would probably like to keep all.of the options open for not all uses are contained in the same analysis. Best wishes, David Booth
BTW in the second paragraph of your question I assumed that independent should be replaced by dependant. Best, D Booth
Although summed scores and factor scores are highly correlated, the later is more appropriate if you want a more precise representation of the construct.
Hi Ceren,
I hope you are doing well! I just want to add that many researchers tend to dislike factor scores as the results are extremely hard to replicate because EFA loadings tend to fluctuate greatly across studies. It is just something to keep in mind :)
Best,
Bryant
- Badges
- Science topic
- Similar topics
- Mathematics
- Statistics
- Regression