I built a rather complex model, consisting of 34 observed variables, 13 latent variables and 3 higher-level latent variables ("areas" or "perspectives" that consist of the lower-lever latent variables). It is similiar to the model "C" from the added link.
I used the SEPATH module of StatSoft's Statistica (v.12 if I remember correctly) to estimate the model parameters using data from my research. I know that the model is too complex for my own good, but it has to be (it is for my PhD thesis, after all) and N=3608. The software calculated the results, the fit indices are pretty good, and suprisingly all parameters (and there are LOTS of them) are estimated with high significance (p
Thanks for your response, Mr. Chalamalla Srinivas, however while the papers you suggested do clarify what the "zeta" represents, it does not quite answer the question, which is how to interpret it's insignificance. Still, jugding from the papers, I am getting closer to believe that there simply is no error in the theoretical assumption of the construct - but I will wait to see what other researchers have to say, just to be sure.
- Badges
- Science method
- Similar topics
- Mathematics
- Statistics
- Statistical Techniques
- Structural Equation Modeling