Hello, I am developing a scale. I prepared the items based on a hypothesis under three factors.
I applied for a pilot study and after that move on to the main study.
I get confused because some sources suggest directly SEM and CFA but others suggest first EFA then SEM and CFA.
Could you please inform me about that title or suggest some sources, please?
Ilhan Albulut, if you are pretty confident that the items you have created accurately represent the three intended factors, I think you could use CFA first up. If that CFA is unsuccessful, I would use EFA to try to identify what "went wrong".
All the best with your venture.
Ilhan Albulut, if you are developing your own scale from scratch, then you need to examine the factor structure using only EFA in the first and second pilot study. Once you start getting a readily interpretable factor solution, you can submit your hypothesised congeneric factor structure to CFA. When your CFA works out well, you can move on to SEM in which you can examine relationships between your factors and other variables of interest.
Article Exploratory Structural Equation Modeling
Article Exploratory Structural Equation Modeling: An Integration of ...
Article Confirmatory Factor Analysis (CFA), Exploratory Structural E...
Ilhan Albulut, I notice you seem not to have appreciated my input up above, but I'll continue to try to help you nevertheless. Unless I misunderstood your "question", when creating your items you seem to have had a clear theoretical notion about the structure of what you intended to measure. Because of that I suggested using CFA first.
The following selections, from page 603 of Hair et al. (full reference below) might be helpful:
"CFA is applied to test the extent to which a researcher’s a-priori theoretical pattern of factor loadings on prespecified constructs (variables loading on specific constructs) represents the actual data. . . . CFA statistics tell us how well our theoretical specification of the factors matches reality (the actual data). In a sense, CFA is a tool that enables us to either “confirm” or 'reject' our preconceived theory."
Here is the reference:
Hair, J. F., Jr., Black, W. C., Babin, B. J., & Anderson, R. E. (Eds.), Multivariate data analysis. Harlow, UK: Pearson.
I hope that's helpful. The suggestions that you move between CFA and EFA, in my view, are good ones.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling