Thank you so much for the informative video. Do we do this between two samples as well? Suppose you are cross-validating the scale in another sample (different country) with a different sample size.
hello sir, you explained the concept very easily. My question is this can be used in longitudinal studies and we have 3 time intervals and 3 constructs in our model?If yes, how . Please suggest videos.
Your videos are so helpful in my study. Thank you so much. I hope you can do some videos on the other 3 tests of invariance. Again, you have done well with the first 2. If I may ask, don't we report the SRMR in the configural and matric invariance as part of model fitness?
thank you so much for your vid sir. I am doing a comparative analysis between two different countries. I followed all the steps you have mentioned in the vid. In configural Invariance in the model fit I've got good model fit with the rest but my AIC was 1513.618. Is it too bad? and also do we need to perform metric invariance as well? coz my model comparison showed significant.... is there other way to proceed with comparative study? if so, please let me know. Thank you so much for the vid.
Thanks. I am glad you liked it. If full invariance is not supported, the researcher can systematically “free” the constraints on each factor that have the greatest impact on chi-square to achieve a non-significant invariance test. If you have a significant chi-square invariance test, you need to see what indicators/factor loadings are differing across the groups. You can look at the individual factor loadings for each group and see where you have large differences across the groups. It may just be one construct causing the significant invariance test. Sometimes the differences in loadings are minimal and give you little direction in finding the indicator that is not invariant across the groups. The best way is to perform an individual chi-square difference test on the indicators of each construct. It is a little tedious but is the best way to see exactly where your measurement problem is compared to just eyeballing differences in factor loadings. For more please refer to the book by Joel E. Collier on SEM using AMOS.
Respected sir, it is quite a nice explanation...however, I wonder if you have a suggestion on how to report measurement invariance...I've read several papers on how to report it, but couldn't get which value to put where and why...Is it possible for you to make a video on how to report and how to conduct scalar invariance, factor variance invariance, and error variance invariance tests...thanks a million
Thanks for your interest. There is a complete series on using AMOS for Structural Equation Modelling. If you can visit this playlist, and refer to video 7, therein i have discussed how to design a model th-cam.com/play/PLb7vm6tsQ3Kvinj9MFF6wbQFDBdkU1ERP.html Hope it helps.
Sir, if some of my configural variance indices are not as per the requirement for model fit, then what to do? My CMIN/df is 1.988, RMR is .068, GFI is .818, CFI is .860, TLI is .840.
Sir, please tell how to do Invariance of Formative Measures (Structure invariance, Slope invariance, and Residual invariance)? Please share a link of the video if u have one...
Hi Dear Professor, If all the groups to be compared are different, then their responses to the questions would also vary, as per their attitude & perception. Then why do we want to establish " Measurement Invariance' before conducting Multigroup analysis?. Although the questions asked are same, but the responses are different for the different groups, so how can they possess measurement Invariance.... Please explain., Sir
Respected professor, when I run the model, it come out with "an error occurred while attempting to fi the model." "an error occurred while attempting to read raw data." what could be the reason for it?
@@researchwithfawad it stated “an error occurred while checking for missing data in the group XXX”, “an error occurred while attempting to fit the saturated model for group XXX, it will not be possible to compute the chi square statistic”, “an error occurred while attempting to fit the independence mode for group XXX, it will not be possible to compute fit measures that require a ‘baseline’ model”. is it the problem from my data file?
Very nice tutorial!
Thanks for watching.
Thank you so much for the informative video. Do we do this between two samples as well? Suppose you are cross-validating the scale in another sample (different country) with a different sample size.
Well I will do it if asked by the reviewer.
Fantastic!!!!
Glad you like it!
hello sir, you explained the concept very easily. My question is this can be used in longitudinal studies and we have 3 time intervals and 3 constructs in our model?If yes, how . Please suggest videos.
Please read more on latent growth curve model using AMOS.
Your videos are so helpful in my study. Thank you so much. I hope you can do some videos on the other 3 tests of invariance. Again, you have done well with the first 2. If I may ask, don't we report the SRMR in the configural and matric invariance as part of model fitness?
Thanks. Hopefully soon. Yes we can report SRMR for fitness.
thank you so much for your vid sir. I am doing a comparative analysis between two different countries. I followed all the steps you have mentioned in the vid. In configural Invariance in the model fit I've got good model fit with the rest but my AIC was 1513.618. Is it too bad? and also do we need to perform metric invariance as well? coz my model comparison showed significant.... is there other way to proceed with comparative study? if so, please let me know. Thank you so much for the vid.
Thanks. I am glad you liked it.
If full invariance is not supported, the researcher can systematically “free” the constraints
on each factor that have the greatest impact on chi-square to achieve a non-significant invariance test.
If you have a significant chi-square invariance test, you need to see what indicators/factor
loadings are differing across the groups. You can look at the individual factor loadings for
each group and see where you have large differences across the groups.
It may just be one construct causing the significant invariance test. Sometimes the differences in loadings are minimal and give you little direction in finding the indicator that is not invariant across the groups.
The best way is to perform an individual chi-square difference test on the indicators of each construct. It is a little tedious but is the best way to see exactly where your measurement problem is compared to just eyeballing differences in factor loadings.
For more please refer to the book by Joel E. Collier on SEM using AMOS.
Great explanation prof. thank you. is this analysis represent the same in PLS(MGA)?
Thanks. assess MICOM when using Multi Group Analysis
Respected sir, it is quite a nice explanation...however, I wonder if you have a suggestion on how to report measurement invariance...I've read several papers on how to report it, but couldn't get which value to put where and why...Is it possible for you to make a video on how to report and how to conduct scalar invariance, factor variance invariance, and error variance invariance tests...thanks a million
Thanks. Hopefully i can make one soon.
at 4:41, you start with the model within AMOS... how do I get there? (linking introductory video would be helpful!) Thank you :)
Thanks for your interest. There is a complete series on using AMOS for Structural Equation Modelling. If you can visit this playlist, and refer to video 7, therein i have discussed how to design a model
th-cam.com/play/PLb7vm6tsQ3Kvinj9MFF6wbQFDBdkU1ERP.html
Hope it helps.
Sir, if some of my configural variance indices are not as per the requirement for model fit, then what to do? My CMIN/df is 1.988, RMR is .068, GFI is .818, CFI is .860, TLI is .840.
This is poor fit. Check item loadings.
Sir, please tell how to do Invariance of Formative Measures (Structure invariance, Slope invariance, and Residual invariance)? Please share a link of the video if u have one...
There is nothing on that yet. I will have to see if same procedure works.
@@researchwithfawad Thanks Prof. I will wait for your able advice and demo nstration through video.
@@researchwithfawad Prof. There is nothing about MIMIC models on your channel..
Hi Dear Professor, If all the groups to be compared are different, then their responses to the questions would also vary, as per their attitude & perception. Then why do we want to establish " Measurement Invariance' before conducting Multigroup analysis?. Although the questions asked are same, but the responses are different for the different groups, so how can they possess measurement Invariance.... Please explain., Sir
Please read about MICOM
Sir, can you please share the Reference for this Configural and Metric Invariance test by this method?
Please refer to this book
Collier, J. E. (2020). Applied structural equation modeling using AMOS: Basic to advanced techniques. Routledge.
Respected professor, when I run the model, it come out with "an error occurred while attempting to fi the model." "an error occurred while attempting to read raw data." what could be the reason for it?
Hi. May be you need to increase the no of iterations. What is the error, any details.
@@researchwithfawad it stated “an error occurred while checking for missing data in the group XXX”, “an error occurred while attempting to fit the saturated model for group XXX, it will not be possible to compute the chi square statistic”, “an error occurred while attempting to fit the independence mode for group XXX, it will not be possible to compute fit measures that require a ‘baseline’ model”.
is it the problem from my data file?
Can you email me the screenshot of the model and the error.
Prof. Can you share the data file used in this video. Thanks.
You can download sample Data here.
I need your email dr if you do not mind.
kh.fawad83@gmail.com