I am a first year PhD student and a new user of MARK and I have a question regarding the POPAN application and RELEASE GOF test. I am trying to obtain yearly population estimates for blackbirds at specific sites in the UK. For approximately 16 years now the birds in these sites have been part of a yearly mark-recapture study, with each year’s study consisting of 12 occasions.
I have read through the MARK manual and after getting stuck have spent time going through chapters 5 and 13 again, but I can’t find any clarification for my problem..... When I run a RELEASE GOF on my fully time dependent models the results are insignificant, indicating that the model is a good fit to the data. But all the tests say there is insufficient data for the GOF test. I know from reading chapter 5 and some of the forum posts here that this problem means that the model can’t be used. I understand that if the lack of fit is extra-binomial then a median c-hat test could be performed instead and the c-hat adjusted in the models. But this test can’t be run in POPAN.
As well as this, the example data used in the MARK manual to explain the application of POPAN also shows insufficient data for the RELEASE GOF run. But this isn’t picked up on in the text, even though a significant GOF test result is. Does this mean that insufficient data for GOF is less of a problem in POPAN application? If insufficient data is a problem can anyone suggest where I go from here?
Also, I am having a similar problem to pstevens, where for the majority of the years I run a GOF in POPAN I am not getting results for TEST 3. Is this also because of insufficient data?
Many thanks to anyone able to help me with this problem!
Jo