Hi, All.
Is it possible to conduct a Confirmatory Factor Analyses on dyadic data? I have answers from 113 heterosexual couples (226 people), and I will be happy to use these data in the process of preparing the adaptation of one questionnaire referring to some parenting issues (it contains 20 items). However, I am not sure whether it is a proper way, as these data are not independent. And conducting CFA seperately on men and women is not possible due to too small sample size.
Do you have any thoughts or references on that topic?
Kind regards,
Kamil Janowicz
Yes, there are special CFA and structural equation models available for dyadic data analysis. See, for example:
Ledermann, T., & Kenny, D. A. (2017). Analyzing dyadic data with multilevel modeling versus structural equation modeling: A tale of two methods. Journal of Family Psychology, 31(4), 442-452.
Ledermann, T., & Macho, S. (2009). Mediation in dyadic data at the level of the dyads: A structural equation modeling approach. Journal of Family Psychology, 23(5), 661-670.
Olsen, J. A., & Kenny, D. A. (2006). Structural equation modeling with interchangeable dyads. Psychological Methods, 11(2), 127-141.
Christian Geiser , thank you for these references. I will check them.
Hello, there is no need to analyze men and women separately for CFA, they can be considered together. In future studies, it may be possible to examine whether gender makes a difference. EFA and CFA can also be done with dyadic variables. There are only minor differences. For example, calculating Kuder Richardson instead of Cronbach's alpha coefficient . But EFA and CFA can be done. However, there are publications stating that the sample size should be at least 300 for CFA, and should even be reduced to the range of 300-500 for a good analysis. Maybe you can try increasing the sample size.
- Badges
- Science topic
- Similar topics
- Statistics
- Survey