I have 4 independent variables from service quality's dimensions.
Considering most of research papers that I read, they don't use covariance in path analysis except in CFA. But, If I don't use covariance to the Independent variables, I got bad model fit. It says, "the following variances are negative"
I need your explanation. Thank you.
In general, the sign of the covariance shows the tendency in the linear relationship between the model's variables. I used the contrivance between the independent variables. please see the attachments.
Dear AMRI, first check the model fit without covar, if yes thesn ok no need to draw covar. if ur model is not fitted, look for Modification indices and draw covar between IVs to fit model. This is usual practice I ever doing. Yes you can draw covar if they are highly correlated with each other.
Thank you so much Prof. Abdelhalim al-zu'bi and Anwar Shah for your thoughts.
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers