You are here

GFI Interpretation using TSSEM

3 posts / 0 new
Last post
VipinSaini's picture
Offline
Joined: 12/24/2019 - 01:53
GFI Interpretation using TSSEM
AttachmentSize
Plain text icon Rscript2.74 KB
Plain text icon Dataset5.04 KB
Plain text icon Output6.93 KB

Dear Mike and Colleagues,

I have conducted a mediating effect MASEM using the random-effect model of TSSEM, and what is surprising me is that R2 is coming 78% and RMSEA is 0.0755 but I am not able to understand how to interpret rest of the indices (SRMR, TLI, CFI) which are going out of range.

I also read a past thread of Prof. Mike which states that the functions of tssem2() and wls() are rewritten (https://openmx.ssri.psu.edu/node/4446). I don't know what is their effect on GFI or how do we deal with it. Is it ok to report them or any alternate to report the data fit with model?

Below is the result of Goodness-of-fit indices:
Value
Sample size 22604.0000
Chi-square of target model 1039.6692
DF of target model 8.0000
p value of target model 0.0000
Number of constraints imposed on "Smatrix" 0.0000
DF manually adjusted 0.0000
Chi-square of independence model 2271.8024
DF of independence model 15.0000
RMSEA 0.0755
RMSEA lower 95% CI 0.0717
RMSEA upper 95% CI 0.0794
SRMR 0.3712
TLI 0.1429
CFI 0.5429
AIC 1023.6692
BIC 959.4621
OpenMx status1: 0 ("0" or "1": The optimization is considered fine.
Other values indicate problems.)

I am also sharing the Rscript, dataset, and output in which I have three independent variables (Ql, SA, RC), two mediators(IC, IH), and one final dependent variable (IA).

Thank you for your guidance!

Mike Cheung's picture
Offline
Joined: 10/08/2009 - 22:37
Your predictors (x1 to x3)

Your predictors (x1 to x3) are forced to be uncorrelated. It may have a negative impact on the model fit.

Mike

VipinSaini's picture
Offline
Joined: 12/24/2019 - 01:53
Thank you Mike, I corrected

Thank you Mike, I corrected the code and now it is running Ok.

Vipin