Message/Author 

Konstantin posted on Wednesday, October 15, 2014  5:12 am



Dear friends, could you help me understand. i have the next output ChiSquare Test of Model Fit Value 18.277 Degrees of Freedom 1 PValue 0.0000 at the same time i have the next statistics CFI/TLI CFI 0.975 TLI 0.772 RMSEA (Root Mean Square Error Of Approximation) Estimate 0.127 90 Percent C.I. 0.080 0.181 Probability RMSEA <= .05 0.004 ChiSquare Test of Model Fit tells me that is not good fit P<0,05 but clf =0,975 and RMSEA <0,05 that telles me about acceptable ﬁt. so what's true? why values of CLF and RMSEA CONFLICTING with plevel of chisquare 


CFI is often much friendlier than Chisquare and RMSEA. I would consider the sample size (which you don't give). If it is very large (several thousand cases) I can imagine going with CFI. At least if the correlations are not very low in which case CFI can mistakenly point to good fit. I would also say that the chisquare and RMSEA results indicate that it is possible to modify your model to get better fit. You can check this out using Modindices. 

Konstantin posted on Thursday, October 16, 2014  2:24 am



yes i have big data set, more than 1000 observations here its https://www.sendspace.com/file/z4ayk4 i have uploaded on web share, cause i can't see how attache file here. programm could work with variables X4X6 so they were excluded 


I have no more comment. We don't want to look at data via Mplus Discussion. 

Konstantin posted on Thursday, October 16, 2014  10:48 am



ooh i'm sorry. but can you help me if i got that result DATA: FILE = "D:/2.dat"; VARIABLE: NAMES ARE y1 x1x6; MODEL: y1 ON x1X6; Observed dependent variables Continuous Y1 Observed independent variables X1 X2 X3 X4 X5 X6 Estimator ML Information matrix OBSERVED Maximum number of iterations 1000 Convergence criterion 0.500D04 Maximum number of steepest descent iterations 20 Input data file(s) D:/2.dat RMSEA (Root Mean Square Error Of Approximation) Estimate 0.000 90 Percent C.I. 0.000 0.000 Probability RMSEA <= .05 0.000 CFI/TLI CFI 1.000 TLI 1.000 what means CFI=1 , is this pretty model or it some error? 


It looks like you have a justidentified model, that is, zero degrees of freedom. This is the case for regression modeling. Such models do not give overall tests of model fit. You will enjoy reading through one of the introductory SEM texts to get more familiar with these issues. 

Back to top 