Back to PLS HelpPosted on 01/29/15 10:03:46
Number of posts: 3
Good morning PLS experts!
I was hoping to get some advise on how to model the interaction of two categorical variables. I came across the following thread that seems to have a similar problem http://www.rotman-baycrest.on.ca/index.php?action=view_thread&id=836&module=bbmodule&src=%40random45c35fcb17881. However, I wanted to confirm if breaking apart the groups into: Group1Male, Group1Female, Group2Male, Group2Female in fact tests for the interaction, Group x Gender ( similar to what we would do in a univariate analysis i.e. SPSS) or if there is a better way when the interaction is the primary hypothesis.
Thank you in advance for your help
Ekaterina
Replies:
Posted on 01/29/15 10:49:33
Number of posts: 394
Good morning PLS experts!
I was hoping to get some advise on how to model the interaction of two categorical variables. I came across the following thread that seems to have a similar problem http://www.rotman-baycrest.on.ca/index.php?action=view_thread&id=836&module=bbmodule&src=%40random45c35fcb17881. However, I wanted to confirm if breaking apart the groups into: Group1Male, Group1Female, Group2Male, Group2Female in fact tests for the interaction, Group x Gender ( similar to what we would do in a univariate analysis i.e. SPSS) or if there is a better way when the interaction is the primary hypothesis.
Thank you in advance for your help
Ekaterina
Hi Ekaterina,
If you parse your data the way you specify, then indeed the PLS analysis will test for the interactions of groupXgender
Posted on 01/30/15 11:40:02
Number of posts: 3
Great, Thank you Randy
Posted on 03/02/15 17:01:14
Number of posts: 3
Just a quick follow-up, is there a way in PLS to control for the main effects of these two factors while examining their interaction?
Thank you for your help
Ekaterina
Login to reply to this topic.
