Dear researchers, can both EFA and CFA be applied to the database obtained at the end of the scale development process at the same time?
Hello Ömer,
I'm uncertain as to what you mean by "...at the same time."
If you are uncertain as to the underlying structure for a set of variables, and have no prior evidence, claim, or theory as to what that structure should be, then EFA is certainly a viable option for exploring and making initial judgments as to what factor space might be appropriate.
Once you determine a defensible candidate structure, it is certainly a good idea to verify that the structure can adequately account for observed relationships among the variables by using CFA with data not used to derive the EFA.
The reason for not applying to the very same data as used for the EFA is the concern that the EFA process can be opportunistic, resulting in a solution which overfits the data set from which it was derived, and therefore may not generalize.
You can find a number of published studies in which author/s have taken part of data set to derive a candidate structure via EFA, then evaluated that structure, via CFA, with the remainder of the data set.
If this doesn't address your concern, perhaps you could elaborate.
Good luck with your work.
Hello dear David Morse Thank you for answer. I've heard that once a scale has been developed, DFA on the initial data to be obtained is flawed. Because CFA shows the fit of a previously tested scale. However, the newly developed scale has not been tested before. Therefore, CFA should not be performed for the newly developed scale. As a result, it is exactly a question mark to apply both EFA and CFA to the collected data for the newly developed scale. However, I could not find a reference to confirm this information. Can you provide a reference to support what I wrote?
Hello Dear Robert Trevethan tahnk you for answer. Suppose you are developing a scale, EFA was performed in the pre-test (pilot) study and would it be correct to perform CFA in the next application? Because in practice, it is imperfect to make both EFA and CFA again. I have question marks about this.
Given a sufficiently large total sample, you could randomly divide it into an exploratory subsample and a confirmatory subsample. Normally, one should make these of roughly equal size. The EFA would both inform the CFA, notably in the absence of compelling theoretical insight, and avoid the CFA's capitalizing on chance from the EFA results because the two sets of results would be from independent samples.
I would precede the EFA with a parallel analysis because selecting the right. number of factors to extract and rotate is the most important decision one makes in conducting a factor analysis.
Omer Sari, I also want refrences for the same issue. Have you find any references related to applying EFA and CFA to different set of data for validation of a new scale? And is the data divided equally or is there any set percentage referenced in any article or book?
- Badges
- Science topic
- Similar topics
- Engineering
- Modeling