Measurement Invariance in Mplus

[@monicajohnson530]

This was extremely useful! Thank you for showing the syntax that simplifies having to run the various tests into one computation.

[@serlide]

Un grand merci pour ce tutoriel et surtout pour avoir mis à la disposition des étudiants les matériels didactique.

[@mariford4667]

Thank you, Llewellyn. This was very useful and easy to follow.

[@senezde]

This is really helping me, thank you so much for explaining so neatly and for providing the material as well. All the best!

[@jacekbuczny4567]

Excellent! I did not know that they created the special comments to test the invariance. Thank you for sharing.

[@MplusforDummies]

Its alot easier now; but the commands dont work well with Longitudinal Invariance. Im busy creating a tool to generate code for LMI at the moment.

[@jacekbuczny4567]

Did you have time to finalize the code for the longitudinal invariance?

[@murex93]

hello, a tutorial on calculating the invariance in cfa with second-order factors would be extremely useful. Are you planning to publish a tutorial about it? Thank you!

[@elliekelly3331]

Excellent explanation, thank you for sharing!

[@LlewellynVanZyl]

Anytime. I have more I depth videos coming on MI. Keep your eyes open 😊 Thank you for taking the time to comment

[@deborahsantos6849]

Very helpful explanation, thanks!

[@MplusforDummies]

Glad this could help you a bit. : )

[@soehartosoeharto8471]

I hope the mimic or DIF analysis in MPLUS can be the new content in this channel, thanks you prof

[@MplusforDummies]

Thanks for the suggestion, I have DIF testing on my list, but MIMIC modelling isnt really my area :(

[@soehartosoeharto8471]

@@MplusforDummies thanks, I will wait for DIF as well. Usually I perform DIF analysis using Rasch modelling

[@MplusforDummies]

@@soehartosoeharto8471 It is alot easier to do in Mplus than in WinSteps. Let me see what I can do. Ill try and post a brief tutorial by the end of December.

[@johnweng8623]

Thank you so much for this - this is wonderful. Question - if the two groups do not have good fit on CFI/TLI - would I just claim that they are invariant at the factorial level and stop further test?

[@ilmiamalia5250]

This video is very helpful, but the link for download does not work

[@derLadenhueter]

thank you for this video. it is really rare to achieve full invariance. how do i interpret partical invariance then? Is there any sense in doing so?

[@fmunozt]

Excuse my English. Thank you very much for the explication. What MPLUS command to use to check "strict invariance"?

[@LlewellynVanZyl]

Its just an additional constraint you add on the scalar model. You can check any of Rens van der Schoot's papers. He usually has nice supplementary material with syntaxes

[@yuheiinoue7616]

Thank you for posting this. I have Mplus 7.0 which does not have the option for configural metric scalar. Could you let me know if there is a way to perform the measurement invariance test without this option?

[@LlewellynVanZyl]

You have to specify it manually then. Here is the syntax for MultiGroup MI. This should get you on your way: www.lesahoffman.com/CLP948/CLP948_Example07a_CFA_MG_Invariance.pdf

[@yuheiinoue7616]

@@LlewellynVanZyl This is great. Thank you so much!

[@RoxyGirlXoXoLovBruno]

How do you run the Wald Test with a modified model? eg. Model: IMAGIN by AQ3 AQ8 AQ14 AQ20 AQ21 AQ24 AQ40 AQ41 AQ42 AQ50; AQ8 WITH AQ3; AQ50 WITH AQ40; AQ21 WITH AQ20;

[@stephscro]

Do you have any resources on how to do this if you are using multiple imputation. The model command configural metric scalar is not available with Mplus if using imputation. Thank you!

[@jacekbuczny4567]

Do the new commands to test the invariance work for other estimators, e.g., MLR?

[@MplusforDummies]

Are you referring to the Scalar Metric Configural Command? If so, then yes. It does

[@h.ozlemanadol6383]

Hi! What level of invariance is ok when I carry out a longitudinal study?

[@MplusforDummies]

It depends on what your research question is. If you want to look at mean differences over time (if that is the overall outcome), then full invariance is required.

[@serlide]

[@rebeccaa6773]

Do you have a video or code for longitudinal measurement invariance? We have repeated measures

[@LlewellynVanZyl]

I just made one on the channel. My latest video. Thanks for the suggestion