There is something off in your SEM output, a TLI & CFI = 1.00 indicate a perfect fit (which is exceedingly rare), but your values are > 1, which is impossible! Are your data normal? Non-normality will mess up your SEM fit indices big time, also a RMSEA of 0.0000 heavily implies some sort of overfitting/saturation. Your chi-square is also through the roof, before you look into the discrepancies between the TAM and the SEM you should figure out what is making your SEM fit indices run wild I think
Which is you sample size? CB-SEM only works with big samples, the requirements for the size depend on the number of variables included in your model. As Marius Ole Johansen said you should also check the distribution of the data.
Natalie no Lam It's small. with 4 or 5 variables in the model we need between 250-300 observations to not have problem with estimating model's parameters.
- Badges
- Science topic
- Similar topics
- Chemistry, Organic
- Organic