I'm doing a validation study and for construct validity, I'll analyze my data through EFA and CFA. So, the same data should be used in both processes or should we divide the data? And if we divide the data what percent of the data should be included in EFA and what percent in CFA?
Given that this is a validation study, you probably won't need to conduct an EFA at all. Typically, in a validation study you would have a priori hypotheses about the number of factors as well as the loading pattern (which variables are supposed to measure which factor(s)). You could therefore proceed with just CFA. I would apply EFA only if you have no idea what the variables are measuring.
Christian Geiser I think that your statement, "I would apply EFA only if you have no idea what the variables are measuring" is too strong. Classic use for EFA is to find out if a set of variables measure just one factor, and can be an open-question for new scales without prior validation.
Of course, a lot depends on what the original author meant by "doing a validation study." If it is based on prior validated scales, then I certainly agree with you about skipping EFA. But if the goal is to use the CFA to validate the EFA, that is another matter entirely. (And I would divide the sample 50/50 in this latter case.)
Thank you David. And yes, I'm not using a pre-validated questionnaire. It is a newly deveopled questionnaire so EFA and CFA both will apply. The only issue is I'm not getting any reference of diving the data for EFA and CFA, and in which percent it should be divided is also not mentioned in any article. So, if possible kindly share any reference.
I am not aware of any citation on this, but a 50-50 split matches a common sense approach.
As mentioned earlier, it is not recommended to use the same data for EFA and CFA. As far as dividing the dataset into two for EFA and CFA is concerned, there are no specific criteria (at least, I did not come across them in the last 13 years). While 50-50 split could be one approach, another could consider the statistical power - Large enough to produce reliable factors in EFA, for example.
The following references will be helpful:
Beavers, A. S., Lounsbury, J. W., Richards, J. K., Huck, S. W., Skolits, G. J., & Esquivel, S. L. (2013). Practical considerations for using exploratory factor analysis in educational research. Practical Assessment, Research, and Evaluation, 18(1), 6.
Kyriazos, T. A. (2018). Applied psychometrics: sample size and sample power considerations in factor analysis (EFA, CFA) and SEM in general. Psychology, 9(08), 2207.
- Badges
- Science topic