Hi,
In order to validate psychometric properties of a questionnaire and more precisely the structural validity, the esem package : https://www.rdocumentation.org/packages/psych/versions/1.9.12.31/topics/esem
indicates tha ESEM is based upon EFA.
Is ESEM necessarely based on EFA ? Or is it possible to base the ESEM on CFA ?
In my case, when I run an EFA, some items converged to the wrong factors. Factors of this questionnaire are supposed to be theoretically highly correlated due to the similarity of certain items. Consequently, if I run ESEM on EFA, I obtain really good fit indexes but with a model which not corresponds to the theoretical model.
Is it possible to conduct ESEM based upon CFA in order to obtain good fit indexes with a model corresponding to the theory? (When I run CFA, fit indexes were quite good but correlation between factors were very high and sometimes superior to 1 because of similarity of certain items).
Thanks
https://tkaiser.science/esemR.html?fbclid=IwAR0HxsOAYCVzDY2V996B4jvjUZx9obQkoFy6mXBr4E8GlHVdj6aP182aKc4
Thanks Cristian Ramos-Vera for this link,
The tutorial indicates that ESEM is based upon EFA corresponding to theoretical expectations. In my case EFA does not correspond to theoretical expectations, so my ESEM is based upon a "wrong" model..
Do the "correlated uniqueness" mentioned in this tutorial allow to specify the model and then run an ESEM based upon an EFA which corresponds to the good theoretical model?
Thanks
Kindly check this publication in RG https://www.researchgate.net/publication/235964237_Exploratory_Structural_Equation_Modelling_ESEM_Application_to_the_SET-37_Questionnaire_for_Students'_Evaluation_of_Teaching
- Badges
- Science topic
- Similar topics
- Indexes