Goodness of Fit Bootstrap v Release

[Lachie Mc]

Dear MARK analysts

When comparing GOF of a 4 occasion CJS "recaptures only" 2 group, time dependent model =phi (g*t) p (g*t) using RELEASE and Bootstrapping methods I get "acceptable" c-hat values ~1.23. However when browsing the results of tests 2 and 3 in RELEASE it tells me that there are insufficient data to perform the test for grp 1. Does this mean I should reject this model as my global model or does the bootstrap resampling regime help account for this meaning c-hat is acceptable and the value 1.23 can be applied to the candidate model set via the adjustment option?

In short the bootstrap tells me things are OK but Release suggests a sparse data set for grp1. Any thoughts much appreciated

Lachie