forum.smartpls.com

Dear all,

I hope someone can help me:)

I am comparing three different groups in smart pls, and have conducted a MICOM analysis. Regarding the analysis, I have two questions:

1. Step 1 and step 2 in the MICOM analysis is fulfilled, and step 3 is partly fulfilled. Does this result mean that I can run PLS-MGA or permutation analysis? Alternatively, does it mean that only a moderation analysis can be run?

2. I have understood that if I run a PLS -MGA with more than two comparisons, I need to take the Bonferroni into account. Do I need to adjust the acceptable p-value in the permutation test and the moderation test as well? Alternatively, can I stick to the standard p-value (0.05) in those tests, when having more than two comparisons?

Thank you some much for your help, I appreciate it!

1. As soon as step 2 is fulfilled you can do a multigroup analysis using PLS-MGA or permutation analysis.
With three groups it is quite complicated to do a moderation analysis (you need two dummy variables representing the second and third category comparing against the first), but you could also do that if you have equal variances.

2. When you do a multigroup analysis with three groups you need to adjust the p-values for multiple testing using Bonferroni or related adjustments. When you do a moderator analysis using dummies you do not need to adjust p-values because you are estimating all moderation effects (i.e., the two dummies) in one model so that the multiple testing is accounted for by the multiple regression analysis in the structural model.

THANK YOU:)