Bias of animal population trend estimates
Edited by: E.J. Wegman, D.T. Gantz, and J. J. Miller
Links
- The Publications Warehouse does not have links to digital versions of this publication at this time
- Download citation as: RIS | Dublin Core
Abstract
A computer simulation study of the population trend estimator used for the Mourning Dove Call-Count Survey, Woodcock Singing Ground Survey, Breeding Bird Survey and other surveys concluded that the estimator had negligible bias in most situations but that observer covariables should not be used with less than five years of data. With rare species (e.g. two birds per route), at least five years should be used. The estimator is seriously biased towards not detecting population changes with very rare species (e.g. 0.3 birds per route). Other technical recommendations are made.
Suggested Citation
Geissler, P., Link, W., 1988, Bias of animal population trend estimates, chap. of Computing Science and Statistics: Proceedings of the 20th Symposium on the Interface, p. 755-759.
| Publication type | Book chapter |
|---|---|
| Publication Subtype | Book Chapter |
| Title | Bias of animal population trend estimates |
| Year Published | 1988 |
| Language | English |
| Publisher | American Statistical Association |
| Publisher location | Alexandria, VA |
| Contributing office(s) | Patuxent Wildlife Research Center |
| Description | xxxvii, 860 |
| Larger Work Type | Book |
| Larger Work Subtype | Other Government Series |
| Larger Work Title | Computing Science and Statistics: Proceedings of the 20th Symposium on the Interface |
| First page | 755 |
| Last page | 759 |