Multi Method with Covariates

questions concerning analysis/theory using program PRESENCE

Multi Method with Covariates

Postby scash » Fri Jun 22, 2012 2:28 am

Hi,

I'm using the single-season, multi-method extension in PRESENCE to compare 4 different survey methods to identify the most efficient technique for detecting each of 9 species of frog. When I've attempted to include detection covariates such as air temperature or rainfall I receive an error message that advises me to check the data file or the design matrix. I've entered the covariate data in the same manner as the detection data (so it should correspond) and set up the design matrix as for other types of analyses. Nichols et al. (2008) makes no mention of environmental covariates of detection and they stick to models looking at the effect of time and method on detection. Is it possible to include environmental covariates for detection probability so that I may compare how these covariates affect detection by the different methods? Or must this be done individually? If it is possible using the multi method extension, does anyone have insight into what I may be doing wrong?

Thank You,
Scott
scash
 
Posts: 3
Joined: Fri Mar 23, 2012 12:37 am

Re: Multi Method with Covariates

Postby darryl » Fri Jun 22, 2012 3:44 am

Hi Scott,
It should all work. Can you fit simple multi-method models with no covariates? What a standard single-season models with covariates for your data set (don't worry if it doesn't make any sense, it's just to check there's nothing funny with your data file)? If these both work, then it could be some other problem with the code, but if you have problems with these, then there's likely something wrong with your data set.
Cheers
Darryl
darryl
 
Posts: 498
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand

Re: Multi Method with Covariates

Postby scash » Mon Jun 25, 2012 2:24 am

Thanks Darryl,

Yes, I can fit simple multi-method models with no covariates and I've just run a single season model with covariates for the data set and it worked as well, so perhaps something with the code as you say?

Cheers,
Scott
scash
 
Posts: 3
Joined: Fri Mar 23, 2012 12:37 am

Re: Multi Method with Covariates

Postby darryl » Mon Jun 25, 2012 5:18 pm

Are you running the latest version of PRESENCE? You can check under the Tools menu
darryl
 
Posts: 498
Joined: Thu Jun 12, 2003 3:04 pm
Location: Dunedin, New Zealand

Re: Multi Method with Covariates

Postby scash » Tue Jun 26, 2012 11:35 pm

I was using version 4.3, then updated to 4.4 but covariates are still not working
scash
 
Posts: 3
Joined: Fri Mar 23, 2012 12:37 am

Re: Multi Method with Covariates

Postby jhines » Thu Aug 23, 2012 12:30 pm

Hi,

Sorry for jumping in late, but were you able to find out why your model with covariates didn't work? If not, I'd be happy to diagnose the problem if you send me the latest zipfile in your project folder.

Jim
jhines
 
Posts: 632
Joined: Fri May 16, 2003 9:24 am
Location: Laurel, MD, USA


Return to analysis help

Who is online

Users browsing this forum: No registered users and 0 guests