MARK analysis

Forum for discussion of general questions related to study design and/or analysis of existing data - software neutral.

MARK analysis

Postby rahultalegaonkar » Mon May 24, 2021 1:22 am

My naïve occupancy results are 0.34, but after analysis in MARK, i am getting my Psi values as 0.95. I have used 8 co-variates. Am i complicating the model?
Also how to extract the site specific values for Psi in MARK?
rahultalegaonkar
 
Posts: 1
Joined: Fri May 21, 2021 6:04 am

Re: MARK analysis

Postby cooch » Mon May 24, 2021 7:44 am

rahultalegaonkar wrote:My naïve occupancy results are 0.34, but after analysis in MARK, i am getting my Psi values as 0.95. I have used 8 co-variates. Am i complicating the model?
Also how to extract the site specific values for Psi in MARK?


Given that you you titled this post 'MARK analysis', wondering why you didn't post it in the MARK-specific sub-forum? No matter -- the questions are general, and not really MARK-specific.

First, your naive estimate of occupancy will always be < true occupancy if p* <1 (where p* is the probablity of the species being detected at least once). So, the fact that your estimates of psi are > naive estimates are expected, and suggest your estimate of p* is <1. In fact, if p* = 0.3579, then given your naive estimate 0.34/0.3579 = 0.95. The number of covariates in the model isn't likely relevant (although you can have problems if some of the covariates are strongly correlated) -- the larger issue is that your detection probability is <1, and thus your estimated psi > naive estimate.

Second, not sure what you mean for site-specific values of Psi -- presumably, you mean the conditional probability a site was occupied even though the species wasn't detected?
cooch
 
Posts: 1628
Joined: Thu May 15, 2003 4:11 pm
Location: Cornell University


Return to analysis & design questions

Who is online

Users browsing this forum: No registered users and 12 guests