Collinear covariates

questions concerning analysis/theory using program PRESENCE

Collinear covariates

Postby loxodonta » Fri Oct 07, 2022 10:51 am

Hello, I have a question about collinear covariates.

I have two collinear covariates, Habitat and Cover, and I'm interested in exploring their effect on occupancy and detection.

Would it be wrong to include Habitat and Cover in the same model if one was specified as a site use covariate, and the other as a detection covariate? e.g., psi(Habitat)p(Cover)?

Many thanks
loxodonta
 
Posts: 2
Joined: Fri Oct 07, 2022 4:19 am

Re: Collinear covariates

Postby jhines » Fri Oct 07, 2022 11:33 am

Yes, occupancy can only be modelled as a function of site covariates, but detection can be modelled as a function of site and/or survey covariates for the standard single-season model.

If you want a model where the 2 covariates are involved with the save parameter, then make sure they are not significantly correlated. In your example, the covariates are on different parameters, so it is OK.
jhines
 
Posts: 598
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA

Re: Collinear covariates

Postby loxodonta » Fri Oct 07, 2022 12:08 pm

Thanks so much!
loxodonta
 
Posts: 2
Joined: Fri Oct 07, 2022 4:19 am


Return to analysis help

Who is online

Users browsing this forum: No registered users and 2 guests