PLS Vid24 PLS Multi Group Analysis

[@prernabagadia4737]

Hello Sir In Multi Group Analysis do we have to evaluate and fulfill the measurement model assessment criteria for all the groups individually.

[@derekong1609]

Prerna Bagadia yes

[@사브리나-u5u]

I did two groups and the results are different males vs females. Now: 1. I should analyze data together then add the analysis for males and females? 2. If i have 200 males and 600 females, is it a fair comparison? 3. If i have foreigners and nationals, but the participants aren't equal. Is it still okay to make comparison ? Is there any rule to follow? Any ref or recommendations?

[@derekong1609]

greetings its not so much of the case of sample size rather than the variance of the two groups... if the two groups variance are comparable then it is okay :)

[@사브리나-u5u]

@@derekong1609 thanks for replying. I noticed that one path wasn't significant in males , females. But it was significant if I merge data. Does it make sense? I worried about it..

[@derekong1609]

@@사브리나-u5u yes it can happen. If you notice the significance of the merge data may not be that high T- value.. then when you separately analyze them.. the significant path should have a higher t-value than the merge data ... and the other can be insignificant

[@parveenaasalam7040]

When we can use parametric test and when PLS/ MGA

[@educationalconsultant9880]

In my model the variable on which I have classified my data in two groups is accreditation and that is secobdary data. Rest all are primary data. Can I run multigroup analysis on that?

[@derekong1609]

my apologies i do not understand your question

[@TankAqil]

Hello, Derek! This is a great video. I have a question about the parametric test table. If the p-value is significant in that table, can we say that gender moderates the relationship? Thanks :)

[@derekong1609]

Yes but its best to use the moderation test to confirm this

[@bilawalakram3775]

dear sir hope you are well i wana know 2 things i have 7 iv,s and 1 mediating variable and 1 dependent variable so ist of all i wanna know for mediation which 2 table will be under observation while interpreting hypothesis test : 1 path coefficient and indirect effects or indirect effects and total effects ???? 2nd is it compulsory to connect all Iv,s with dependent variable in PLS 3.0 IF i connect all Iv,s with Mediating variable and connect mediating with Dv is it fine as u did it in your video ???

[@derekong1609]

1st question... please see my video in this playlist video number 20 for mediation. 2nd question... it all depends on what hypothesis you are intending to find,, that will determine your structural model

[@afaadd4905]

Dear sir hope well. i analyzed my data by following your vids but i faced issued in MGA because it does not analysis indirect specific effect, others words , it does not analysis mediation effect in MGA. My study model contains two mediation hypothesis, which are not measuring by MGA, so what i do. Even you can check it, there is not any specific indirect effect option in MGA.

[@derekong1609]

Greetings does that mean you are unable to find the mediation effect for both the male and female?. Actually may I know what is the reason of finding mediation comparison in MGA and what are you trying to show...as far as I am concern there is no need for that in MGA... if you would like to see if there is mediation for each group may I suggest that you model the mediation for each group separately :) thanks

[@halilcicibas2038]

Thank you for the videos. Do we need to conduct invariance measurement test before MGA?

[@derekong1609]

Yes that would help

[@shinajvs527]

Dear Derek Ong, Is it required to conduct omnibus test of group differences (OTG) if the study compare different categories under one group. For instance, if the study attempts to examine the heterogeneity effect in a single group, namely 'Authority type', and that includes three categories i.e. 1)Government; 2) Semi-government; and 3) Private. Sarstedt, Henseler, and Ringle (2011) proposed the omnibus test of group differences (OTG) for analyzing the differences of two and more groups of relationships. Based on the case mentioned above, according to Sarstedt, Henseler, and Ringle (2011), what is considered as a group? is it 'Authority type' ? or three classication stated under the group? Can i just conduct PLS- MGA in above case?

[@derekong1609]

Greetings please send me a personal email to answer your query

[@shinajvs527]

@@derekong1609 I really appreciate your prompt response. [e-mail]

[@irnalestari6126]

Hello sir, Thanks for the video. I have this moderator variable AGE with 4 groups. And I already do Multi Group Analysis step by step based on your video. But when in MGA calculating, there is only 2 Group to choose. How to frame the hypothesis sir? Can we just do hypothesis test only from bootstrapping result? Thank you sir

[@derekong1609]

You may want to group your moderator into groups in SPSS or Excel and then re read the data again into SMART PLS

[@irnalestari6126]

@@derekong1609Thanks for your reply Sir. I have generated my moderating variable into 4 group in smartPLS. I have run Bootstrapping (Data Group) for this 4 groups but 1 group have singular matrix problem, so the result doesn't come out. And Multi Group Analysis did not allowed to compared 4 groups at a time. But when i try to compare 2 by 2, it has also singular matrix problem because the group does not have similar sample size. Is there any method to test the hyhothesis in this case? Thanks in advance

[@derekong1609]

@@irnalestari6126 Singularity Matrix problem happens when you have near perfect correlation for one of your groups. You might want to consider regrouping or add cases. if you run the correlation you will find that some of your groups have near perfect correlation.

[@prernabagadia4737]

Hello Sir How to frame the hypothesis for MGA if i am having moderating variable with three sub groups

[@derekong1609]

It depends what is your variable

[@prernabagadia4737]

Greeting sir The variable is independent variable named financial risk tolerance having three category low medium and high. We want to show the effect of financial risk tolerance on the path model having six exogenous construct and one endogenous construct

[@derekong1609]

@@prernabagadia4737 you might have to do a comparison 2 by 2 ...as MGA does not do more than 2 categories....in some cases some variables can be used as quantitative measure

[@prernabagadia4737]

Hello Derek Sir Will u please guide me how we can fulfill the measurement model assessment criteria for all the groups. In my model I am having three groups and all the groups varies in their assessment criteria. Fulfilling one group assessment criteria by deleting one of the item will make changes in other group which is not required for that group.

[@derekong1609]

Prerna Bagadia then just look at the main assessment model

[@prernabagadia4737]

Hello sir The main assessment is all ok. But when we look the assessment model group wise then for some value it comes below the standardised value

[@derekong1609]

Prerna Bagadia keep it to main assessment

[@prernabagadia4737]

Sir This means that the GroupWise Assessment is not required in my model. Will it not give any effect in my further analysis as groupwise assessment is not getting to the mark?