Calculating p*(cumulative pdet on a site over all K surveys)

questions concerning analysis/theory using program PRESENCE

Calculating p*(cumulative pdet on a site over all K surveys)

Postby Lea » Wed Jul 26, 2017 10:34 am

This is a pretty basic question but I just wanted to make sure I was calculating P* correctly (P*=(1- (single survey p)^K)). Am I correct in assuming that the single survey p is the Individual Site estimates of p from my model output? Looking at the Blue-lined salamander example it seems so.
Lea
 
Posts: 30
Joined: Thu Oct 24, 2013 1:09 pm

Re: Calculating p*(cumulative pdet on a site over all K surv

Postby jhines » Wed Jul 26, 2017 10:40 am

That is correct for the model where the single-survey p does not change by survey (model p(.)). A more general formula is:

p* = 1 - product( 1-p(j)) {j=1,2,... K}

where p(j) is the detection probability for survey j.
jhines
 
Posts: 386
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA

Re: Calculating p*(cumulative pdet on a site over all K surv

Postby Lea » Wed Jul 26, 2017 11:50 am

Thank you!
Lea
 
Posts: 30
Joined: Thu Oct 24, 2013 1:09 pm

Re: Calculating p*(cumulative pdet on a site over all K surv

Postby darryl » Wed Jul 26, 2017 4:20 pm

Hi Lea,
Note that your initial equation of:
P*=(1- (single survey p)^K)

isn't quite correct. It should be

P*=1- (1-p)^K

where p is the single-survey detection probability. Your P* would be the probability of not detecting the species in every survey (i.e., 1 minus the probability of detecting it in every survey), given presence.

Darryl
darryl
 
Posts: 405
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand


Return to analysis help

Who is online

Users browsing this forum: No registered users and 1 guest