Keyboard Shortcuts
ctrl + shift + ? :
Show all keyboard shortcuts
ctrl + g :
Navigate to a group
ctrl + shift + f :
Find
ctrl + / :
Quick actions
esc to dismiss
Likes
Search
#help How to deal with collinearity between independent variables when running MRQAP ?
#help
|
Hi all,
I encountered a problem when running MRQAP. I try to investigate which factors influnce formal collaboration between cities(weighted and undirected). I have two independent variables: IV1-informal interaction (weighted and undirected); IV2-network of prescribed interaction (binary and unweighted), and one control variable. When I? include IV1 and the control variable, IV1 is insinificant and sign is +. However, when IV2 is included, IV1 is significant and the sign is -.? ?I guess that issue may arise due to colinearity between IV1 and IV2. Is there a solution like? the residualization strategy in OLS to sovle this problem? (maybe it sounds strange¨s¦ä¨t)
Any information would be appreciated. Thank you!
?
Best,
?
Longxia Huo
?
Ph.D. Student
|
|
Hi, Assuming no violation of other OLS assumptions, you can go ahead as you choose. However, I wouldn't dismiss the outcome or see it as a problematic?outcome,?necessarily. By controlling for the prescribed network, you are in fact picking up formal collaboration and communication associated with it. Informal interaction may be less needed, because there is formal interaction already, while on the other hand when there is no formal collaboration, there will be more informal interaction needed. Good luck. Best regards, David Op vr 4 nov. 2022 om 11:12 schreef Longxia Huo <15927173853@...>:
|
to navigate to use esc to dismiss