15. Measurement Invariance of Composites (MICOM) in SmartPLS-4 || Dr. Dhaval Maheta

#sem, #smartpls, #construct, #latent, #model, #micom, #configural, #compositional, #partial, #full, #measurement, #variance, #micom, #smartpls4
Email: [e-mail]
Twitter: / dhavalmaheta77
LinkedIn: / dhaval-maheta-320200153
Facebook ID: / dhaval.maheta.37
Telegram ID: t.me/DhavalMaheta

[@karanmehta1987]

I've been looking for a detailed video on MICOM for quite a while and this video is perfect for my paper's analysis. Thank you for your guidance sir.

[@user-so4gn6wm2h]

你太厉害了 感谢感谢！！

[@user-se4ve7sx5z]

Thank you very much for the informative video. Just an opinion, it would have been much better to show how the values should be reported in a tabular format. In addition, there are some research articles that recommended additional tests while doing MGA with more than two groups. Example, "Multigroup analysis of more than two groups in PLS-SEM: A review, illustration, and recommendations."

[@DhavalSaifaleeAaryash]

You are right

[@dishachoudhary9261]

Thanku Sir,my partial measurement is established ..I would want to know how to report this in research paper. I m also analysis MGA among male female..after running MGA ...we should we report bootstrapping results sir only what if we have meditaing path also..

[@dishachoudhary9261]

Plz reply

[@chefberrypassionateresearcher]

Sir, please tell how to do Invariance of Formative Measures (Structure invariance, Slope invariance, and Residual invariance)?

[@chefberrypassionateresearcher]

u have not shown step 3 of micom. if i dont see any permutaion p values (blank) , then what to do?

[@chefberrypassionateresearcher]

Sir, if i want to do one way ANOVA, shall i first do measurement invariance test, so that i am able to get a model with valid indicators for my contructs (as i will remove those with inadequate loadings in second step of MICOM) & then do the descriptives and one way ANOVA?

[@anugrewal9790]

Greetings sir if I have age , gender, qualifications as my control variable then I will perform analysis through dummy variables? Do I need to use micom ,mga?

[@shagunsarraf2312]

Hi, Thank you for such a great video. I have one question in my step 2 results all values are above 0.05, but only one value is less than 0.05 but greater than 0.01. Can I consider proceeding to step 3?

[@DhavalSaifaleeAaryash]

Yes

[@SIA_0797]

Very interesting video. We run a cross-country MGA, and the results are awesome for the complete model (construct validity, discriminant&p-values), but not so great for each country. What are we missing? Many thanks.

[@DhavalSaifaleeAaryash]

You are not missing anything. It means that in some paths relationships are universally true...

[@Alhamzah_F_Abbas]

Thank you for the video. Is this MICOM consider a necessary step before running MGA, or I can go with MGA directly?

[@DhavalSaifaleeAaryash]

Yes

[@anakagungwirayuda4012]

how to do OTG in SmartPLS?

[@FatmaAhmedRagab]

Sir I followed your steps and I have a partial invariance but I get insignificant p values in permutation test for all constructs and in the MGA bootstraping result I found a difference between the two group. Why is there a difference in my results? and what is the correct results/test to be reported?

[@DhavalSaifaleeAaryash]

If steps are complete and p value is insignificant it means that it not necessary to report separate results

[@cata.maican]

Hi, Could you, please, clarify how the analysis should be performed if the group contains more than 2 values as in your case (gender=M/F) - in the MGA window it is possible only to compare 2 values, so if I want to compare among multiple values, e.g. A, B, C, what should I do? Compare A-B, A-C, B-C and then how the data should be reported? Thank you

[@user-se4ve7sx5z]

I had the same problem. There are very few videos and research papers discuss more than two groups. I think, you can compare e.g. A, B, C, with each two groups: such as A-B, A-C, B-C. For more details: you can read the following article: "Multigroup analysis of more than two groups in PLS-SEM: A review, illustration, and recommendations." It recommended many additional tests while doing MGA for more than two groups.

[@user-uu6tv7mr9l]

Sir what is the minimum number of respondents needed in a group to run multi group analysis in smart pls 4

[@DhavalSaifaleeAaryash]

atleast both groups should have equal representation

[@KaranBerry-ol2hg]

Professor please answer: Q1. Do i need to do MICOM (measurement invariance) as first step whenever i want to find out any statistical differences or relationships like ANOVA, Kruskal Wallis, or even descriptives, as during MICOM second step, if some of the items may be insignificant in my constructs, i will have to remove them... So the final model will not have those items. Sir, please clarify..

[@DhavalSaifaleeAaryash]

No for anova and kruskal wallis test not necessary

[@KaranBerry-ol2hg]

@@DhavalSaifaleeAaryash Sir, but what if i removed some items after doing second step of MICOM.?

[@chefberrypassionateresearcher]

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.

[@DhavalSaifaleeAaryash]

sir we are compaing their demographics. If demographics are same then we will check for the difference in responses.

[@chefberrypassionateresearcher]

@@DhavalSaifaleeAaryash Sir, please clarify further in detail..I dont understand what to do you mean by comparing their demographics.

[@shreyamukherjee9792]

Please give paper reference and is it same for formative and reflective constructs?

[@DhavalSaifaleeAaryash]

It is given in book Partial Least Squares written by Professor Ringle

[@shreyamukherjee9792]

Thank you