MGA results vs Groups bootstrapping

[alonsodossantos]

Hello everyone. I have one dude, please can you help me?

I am testing a multi-group model. Some relationships shows bootstrapping differences. For example:
Quality -> Satisfaction Group 1 p-value 0.236; Group 2 p-value 0.001
BUT the PLS-MGA and others indices shows a p-value of 0.240.
What that means? Is it a difference between groups or not?. I do not understand very well.

Please help.

Manuel Alonso

[jmbecker]

The difference between significant and nonsignificant might iftself be not significant. You can finde many paper about this phenomenon.
Important is only the significance of your difference test and not whether there are differences in significance within your groups.

[alonsodossantos]

Thank you for the reply. Is it possible to have some references about this phenomenon?
Thank you again.

Manuel

[jmbecker]

Just put the first sentence into google and you will find references.