total N differs in derived with factor variables

questions concerning anlysis/theory using program DENSITY and R package secr. Focus on spatially-explicit analysis.

total N differs in derived with factor variables

Postby Ulrike » Thu Jan 16, 2020 10:16 am


This question refers to openCR, I don't know whether a similar issue arises in secr.

I have done an analysis using a factor covariate with 2 different possible values, 0 and 1.
I then use derived function, option ,all.levels=TRUE, to get a nice summary of estimates for capture prob, survival, entry prob etc.
However, I would also be interested in an easily accessible estimate of abundance.
However, this is not my main question.
In derived, a message shows that "counts and abundance estimates refer to all animals". Thus, these should be the same for all factor levels, which is mostly the case, but not for N (however N is clearly in this case the abundance of all animals). For N, the numbers differ for the 2 factor levels. Is there any explanation for this?
Does it mean anything like, if all animals detected were from this group, then N would be the total abundance estimate (for each time point)?

Thanks and best regards, Ulrike
Posts: 5
Joined: Sat Aug 31, 2019 10:24 am

Re: total N differs in derived with factor variables

Postby murray.efford » Sat Feb 01, 2020 6:51 pm

If this is still an issue - can you possibly provide the code for a minimal example? Use one of the builtin datasets and a fake covariate if necessary. I assume you are referring to a nonspatial analysis (derived does not report N for spatial models) and hence there's no equivalent in 'secr'.
Posts: 595
Joined: Mon Sep 29, 2008 7:11 pm
Location: Dunedin, New Zealand

Return to analysis help

Who is online

Users browsing this forum: No registered users and 1 guest