My dataset consists of 120 observations, and 7 observed variables. I'm using 3 factors (result from parallel analysis and theory). All 7 variables load relatively well (2 in factor 1, 2 in factor 2, 3 in factor 3). The Tucker Lewis Index = 1.034 and CFI = 1.004, which looks OK. However, I'm getting RMSEA and RMSR values of zero and I'm not sure what to make of this. Any help is appreciated.
It appears that you have very good fit. Any time that chi-square is less than df, then RMSEA will be zero.
But the value of RMSR is a bit troubling. Is it exactly zero? I would be a little suspicious, and check this out further. Can you get the predicted correlation matrix, and compare it to what you have?
CFI is also troubling. CFI cannot go above 1.00 (normally). (TLI can – that's OK.)
- Solved – do parallel analysis with any type of exploratory factor analysis/principal component analysis
- Solved – How to correctly interpret a parallel analysis in exploratory factor analysis
- Solved – Why eigenvalues are greater than 1 in factor analysis
- Solved – Deleting variables from an exploratory factor analysis
- Solved – How to interpret the fit of cfa (sem) outcome