Please don't misuse the museum: ‘declines’ may be statistical

摘要: Detecting declines in populations at broad spatial scales takes enormous effort, and long-term data are often more sparse than is desired for estimating trends, identifying drivers population changes, framing conservation decisions, or taking management actions. Museum records historic can be available large across multiple decades, therefore an attractive source of information on the comparative status populations. However, changes may real (e.g. response to environmental covariates) resulting from variation our ability observe true (also possibly related covariates). This a (statistical) nuisance understanding population. Evaluating statistical hypotheses alongside interesting ecological ones important appropriate use museum data. Two considerations generally applicable records: first without initial random sampling, comparison with contemporary results cannot provide inference entire range species, second availability only some individuals respond changes. Changes reduce proportion that present able counted given survey event, apparent decline even when size stable.