## Model fit error

dennylarson
PLS Junior User
Posts: 5
Joined: Tue Jul 07, 2020 2:37 am
Real name and title: Denny Larson

### Model fit error

This is my result of using PLS Algorithm of model fit
SRMR 0.113 0.114
d_ULS 2.684 2.721
d_G n/a n/a
Chi-Square infinite infinite
NFI n/a n/a

Why do i get this kind of result
somebody pls help me this is very important and close to deadline report sumbission
Attachments
model fit.jpg (212.44 KiB) Viewed 7497 times
jmbecker
SmartPLS Developer
Posts: 1129
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

### Re: Model fit error

As I have already answered to other posts in this forum, it is not possible to calculate all fit measures for all models. For some models it is simply not possible to calculate NFI and Chi-Square as well as d_G. In most cases, these are models with repeated indicators, for example, when you model higher-order constructs. If you really need the fit measures, built a two-stage model. However, it might also be a good question whether you really need the fit measures? See also https://www.smartpls.com/documentation/ ... /model-fit
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
dennylarson
PLS Junior User
Posts: 5
Joined: Tue Jul 07, 2020 2:37 am
Real name and title: Denny Larson

### Re: Model fit error

jmbecker wrote: Tue Jul 07, 2020 1:46 pm As I have already answered to other posts in this forum, it is not possible to calculate all fit measures for all models. For some models it is simply not possible to calculate NFI and Chi-Square as well as d_G. In most cases, these are models with repeated indicators, for example, when you model higher-order constructs. If you really need the fit measures, built a two-stage model. However, it might also be a good question whether you really need the fit measures? See also https://www.smartpls.com/documentation/ ... /model-fit
I don't understand where do i repeat my indicator, i think all my indicators are different. how do i build a two-stage model
jmbecker
SmartPLS Developer
Posts: 1129
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

### Re: Model fit error

Without knowing your model and data, it is hard to tell the problem
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
dennylarson
PLS Junior User
Posts: 5
Joined: Tue Jul 07, 2020 2:37 am
Real name and title: Denny Larson

### Re: Model fit error

jmbecker wrote: Tue Jul 07, 2020 2:09 pm Without knowing your model and data, it is hard to tell the problem
Does this help ?
Attachments
Screenshot (16).png (66.7 KiB) Viewed 7483 times
jmbecker
SmartPLS Developer
Posts: 1129
Joined: Tue Mar 28, 2006 11:09 am
Real name and title: Dr. Jan-Michael Becker

### Re: Model fit error

The constructs of perceived parental support and self efficacy are so highly correlated (as evidenced by a very large standardized path coefficient), it is very likely that the models suffers from a multicollinearity problem. Maybe even some of the indicators of these constructs are very highly correlated. You might have a look at the correlation matrix as well.
I can also not imagine that these constructs actually show discriminant validity.
Dr. Jan-Michael Becker, University of Cologne, SmartPLS Developer
Researchgate: https://www.researchgate.net/profile/Jan_Michael_Becker
dennylarson
PLS Junior User
Posts: 5
Joined: Tue Jul 07, 2020 2:37 am
Real name and title: Denny Larson

### Re: Model fit error

jmbecker wrote: Tue Jul 07, 2020 3:06 pm The constructs of perceived parental support and self efficacy are so highly correlated (as evidenced by a very large standardized path coefficient), it is very likely that the models suffers from a multicollinearity problem. Maybe even some of the indicators of these constructs are very highly correlated. You might have a look at the correlation matrix as well.
I can also not imagine that these constructs actually show discriminant validity.
What usually cause this correlated problem ?
How can one fix it
malik123
PLS Junior User
Posts: 1
Joined: Sun Nov 22, 2020 6:16 am
Real name and title: malik reddy

### Re: Model fit error

The builds of apparent parental help and self-adequacy are so profoundly connected (as confirmed by an extremely huge normalized way coefficient), all things considered, the models experience the ill effects of a multicollinearity issue. Perhaps a portion of the pointers of these builds are exceptionally corresponded. You may examine the connection lattice too.
dorathylee
PLS Junior User
Posts: 2
Joined: Mon Jan 04, 2021 2:49 pm
Real name and title: Lee lin

### Re: Model fit error

I have three Query about the exact model fit:
1
d_ULS and d_G must smaller than the confidence interval, then the confidence interval that d_ULS and d_G have to compare with is the 95% 99% value obtained from bootstrapping calculated?
Or do we have to calculate based on the path coefficient?

2. The other criterion is that the outcome must be nonsignificant.
If Chi-square is not in the index to determine whether the outcome is significant or not? Does the condition also come from the confidence interval of the Path coefficient? if zero is included in the interval, then is it significant? Or what is the standard to judge?

3. If the value of the saturated and estimated model is the same ,
Some scholars say that it means that the saturated model is fully
fit with the estimated model.
But if the outcome can’t fulfill the significant and d_G ,d_ULS criteria
, can we still conclude that there is a good fit?
dorathylee
PLS Junior User
Posts: 2
Joined: Mon Jan 04, 2021 2:49 pm
Real name and title: Lee lin

### Re: Model fit error

This is my Data
Table 12 Outcome of the model fit test.
discrepancy SRMR D_ULS D_G NFI Chi-square

Saturated model 0.062 0.967 0.458 0.836 1010.938

Estimated model 0.062 0.967 0.458 0.836 1010.938

95% 0.050 0.935 0.442
0.053 1.074 0.446
99% 0.052 1.023 0.476
0.056 1.195 0.477

GOF √0.5417 * 0.4035= 0.4674 > 0.36 Large GOF