Burnham Tests

Burnham et al. gives three types of statistical tests for assessing model assumptions for the release-recapture protocol.

1. TEST 1: Tests for overall treatment effects.

   H₀: All survival and detection parameters are the same across treatment groups.

   H_A: At least some parameters differ between or among groups.

2. TEST 2: Tests for goodness of fit of the model to the data.

   H₀: The survival and detection parameters are specific to sampling occasions within each group.

   H_A: The model does not fit the data; there may be tagging effects or differential behavior.

3. TEST 3: Tests for parameters that are specific to individual capture histories.

   H₀: The survival and detection parameters do not depend on the capture histories of fish released on any rele ase occasion.

   H_A: Some of the parameters are dependent on the capture histories of fish in a given release.

Top of page