thank you very much Dr. Mostafa for this amazing workshop. I have a question I hope I can get an answer from you. the results of my PLS-MGA for gender are all in red, the same results you run in the video. in this case. is there an issue in my model or my groups? because previous literature shows that male and female has different results. how I can interpret the results in this case?. I will be very grateful for your reply.
Dear sir. Greetings, I have a query regarding MICOM analysis. In my study I have considered the trading experience as categorical moderator. As a prerequisite, I checked the MICOM. Out of 9 constructs 2 constructs are failed to fulfill the criteria of compositional invariance. Here . I would like to know. How should I proceed further? Thank you
You need to check the item loadings of problamatic construct from two groups data, and remove an item with highest loading difference between two groups.
@@s.mostafarasoolimanesh7567 I didn't get you sir....will you plz elaborate? Do you mean 1 moderator on 1 relationship? While doing so, shall I remove other relationships from the model..??
Dear Mostafa, I have four IV, one mediator, one DV and one moderator. The moderator is age and I want to use MGA. But in my conceptual framework age only moderates mediator and DV. It's a moderated mediation. If I use interaction effect, age is not a continuous variable so I cannot use it either. If I use MGA, I only want to test one specific path. Literatures says MGA is used to test the whole structural model instead of specific indirect path. So what should I use to test the moderated mediation. Can I still use MGA?
@@s.mostafarasoolimanesh7567 Thank you very much. Prof. Mostafa. I passed my viva finally. Incredibly, three years later, I come back again to find this video. Now this time I focus on interaction effect in my paper. I find running MGA has a higher requirement for the validity and reliability for each group. For example, my HTMT is below 0.9 for the whole sample pool. Yet when it comes to each group, my HTMT is over 0.9. I tried different sample size to fix this issue but I failed. Now I'm thinking about using interaction effect. My moderator is job positions(FOH vs. BOH). Am I understanding it right that MGA has a higher demand for sample data?
See this paper of mine: Rasoolimanesh, S. M., Badarulzaman, N., Abdullah, A., & Behrang, M. (2019). How governance influences the components of sustainable urban development?. Journal of Cleaner Production, 238, 117983.
One video on this how we can analyze this "The combined use of symmetric and asymmetric approaches: partial least squares-structural equation modeling and fuzzy-set qualitative comparative analysis"
Sir, can u share the references for the using of two stage for all types of indicators, both all reflective or mixed with formative. I need it sir. Hopely u can responsme me...thx
Sir I have a question and I hope you will answer it when you have free time. Sir my question is that I have DV ( sustainable performance) and in literature it is a multidimensional construct based on three dimensions that are economic performance, environmental performance and social performance, now its conceptual definition is also based on three dimensions now sir can operationalise this construct based on two dimensions that is environmental performance and social performance and can skip the economic performance.? Reason: sir actually my target organisation are public sector hospitals and mostly they are non profit organisations and there is no economic performance (growth profit, market share etc) in these organisations. Secondly can I run analysis on pls? In this situation or should go for Amos. ? Or can do efa separately and then run pls ? JazakAllah
Sir I only have an argument that i skipped the economic performance dimension because my target Organization is public sector Organization that why I am measuring performance based on two dimensions Environmental and social. Is this valid argument ?
Very informative thank you so much dear Mostafa
Thank u very much Dr
thank you very much Dr. Mostafa for this amazing workshop. I have a question I hope I can get an answer from you. the results of my PLS-MGA for gender are all in red, the same results you run in the video. in this case. is there an issue in my model or my groups? because previous literature shows that male and female has different results. how I can interpret the results in this case?. I will be very grateful for your reply.
Red results mean, there is no significant difference between two effects.
@@s.mostafarasoolimanesh7567 thank you so much for your reply prof.
Dear sir.
Greetings,
I have a query regarding MICOM analysis. In my study I have considered the trading experience as categorical moderator. As a prerequisite, I checked the MICOM. Out of 9 constructs 2 constructs are failed to fulfill the criteria of compositional invariance. Here . I would like to know. How should I proceed further?
Thank you
You need to check the item loadings of problamatic construct from two groups data, and remove an item with highest loading difference between two groups.
Sir can I use moderation separately for all my 7 hypotheses instead of PLS MGA..?
Yes, you can. But, you need to do one-by-one.
@@s.mostafarasoolimanesh7567 I didn't get you sir....will you plz elaborate?
Do you mean 1 moderator on 1 relationship? While doing so, shall I remove other relationships from the model..??
@@Shravan1231000 No, you should not remove other relationships. Keep all relationships, but include moderator for each relationship and run the model.
Dear Mostafa, I have four IV, one mediator, one DV and one moderator. The moderator is age and I want to use MGA. But in my conceptual framework age only moderates mediator and DV. It's a moderated mediation. If I use interaction effect, age is not a continuous variable so I cannot use it either. If I use MGA, I only want to test one specific path. Literatures says MGA is used to test the whole structural model instead of specific indirect path. So what should I use to test the moderated mediation. Can I still use MGA?
Yes, you can use MGA to test the differences between one path only.
@@s.mostafarasoolimanesh7567 Thank you very much. Prof. Mostafa. I passed my viva finally. Incredibly, three years later, I come back again to find this video. Now this time I focus on interaction effect in my paper. I find running MGA has a higher requirement for the validity and reliability for each group. For example, my HTMT is below 0.9 for the whole sample pool. Yet when it comes to each group, my HTMT is over 0.9. I tried different sample size to fix this issue but I failed. Now I'm thinking about using interaction effect. My moderator is job positions(FOH vs. BOH). Am I understanding it right that MGA has a higher demand for sample data?
Dear Mostafa, could you please let me know how to examine secondary data on smartPLS? thank you
See this paper of mine:
Rasoolimanesh, S. M., Badarulzaman, N., Abdullah, A., & Behrang, M. (2019). How governance influences the components of sustainable urban development?. Journal of Cleaner Production, 238, 117983.
@@s.mostafarasoolimanesh7567 Noted thank you for your cooperation. I will be in touch with you.
Respected Dr would you advise me how to improve the moderator ? Thx u
What do you mean, advise to improve moderator? Please watch moderator videos, you can get useful insights.
One video on this how we can analyze this "The combined use of symmetric and asymmetric approaches: partial least squares-structural equation modeling and fuzzy-set qualitative comparative analysis"
Will upload soon!
Dear Sir, only up to Week7? no more in Week8?tq
Week 7 is the last week.
Sir, can u share the references for the using of two stage for all types of indicators, both all reflective or mixed with formative. I need it sir. Hopely u can responsme me...thx
Sir can we test theory through smart pls ? Or we should go for amos?
Yes, you can. The research can be exploratory or confirmatory. But for pure theory testing the CB-SEM is recommended.
Sir I have a question and I hope you will answer it when you have free time. Sir my question is that I have DV ( sustainable performance) and in literature it is a multidimensional construct based on three dimensions that are economic performance, environmental performance and social performance, now its conceptual definition is also based on three dimensions now sir can operationalise this construct based on two dimensions that is environmental performance and social performance and can skip the economic performance.? Reason: sir actually my target organisation are public sector hospitals and mostly they are non profit organisations and there is no economic performance (growth profit, market share etc) in these organisations. Secondly can I run analysis on pls? In this situation or should go for Amos. ? Or can do efa separately and then run pls ? JazakAllah
If you have strong justification for excluding the economic performance, you you can.
Sir I only have an argument that i skipped the economic performance dimension because my target Organization is public sector Organization that why I am measuring performance based on two dimensions Environmental and social. Is this valid argument ?