Handling sparse data for JMV GOF test

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

Handling sparse data for JMV GOF test

Postby sharrod » Thu Feb 28, 2019 4:04 pm

Hi everyone,

I am analyzing recruitment (probability of a juvenile surviving and being "recruited" into the breeding population) of a banded bird population. I have 16 encounter occasions, and my states are as such: 2=banded as a chick, 1=chicks recaptured the next season as breeding adults. I'm conducting GOF tests on my data in UCARE, but for most of these, I get values of 0 or NaN. Based on the last post of this discussion thread (http://www.phidot.org/forum/viewtopic.php?f=5&t=1051), and my own observations of the encounter histories file (see sample below), it seems that the problem is originating from having too few individuals in many of the capture histories.

Excerpt from encounter histories file:
0000000000000002 509 ;
0000000000000020 602 ;
0000000000021000 15 ;
0000000000021000 -1 ;
0000000000021100 10 ;
0000002000010000 1 ;
0000002001110000 1 ;
0000002100000000 1 ;
0000002101000000 2 ;
0000000000021111 1 ;
0000000000200000 479 ;

That being said, I haven't been able to find any solutions for how to deal with this problem. I've looked through the UCARE manual, other phidot forum threads, and multiple other sources that mention multistate analyses, but have come up empty. Does anyone on here know of how to deal with this issue? All I can think to do is remove some of the encounter histories with only 1 or 2 individuals, but that seems like it would be skewing the data. Any ideas or suggestions would be appreciated.

Thank you,

Sara
sharrod
 
Posts: 10
Joined: Thu Aug 24, 2017 12:05 pm

Return to analysis & design questions

Who is online

Users browsing this forum: No registered users and 14 guests