U.S. Department of Agriculture: Animal and Plant Health Inspection Service
Date of this Version
December 1993
Abstract
Biodiversity is the bandwagon of the moment. Unfortunately, the concept is sufficiently complex that almost any population biology study, with almost any conclusion, can be framed as an effort to measure or conserve biodiversity. Based on what 1 have seen in the literature and heard at recent scientific meetings, here is a primer on some of the more popular ways to bend biodiversity data. Suppose you wish to claim that a species is disappearing. With the explanation that time and funding were limited, you might present population trajectories based on as few as two estimates of abundance. Perhaps the final abundance estimate was obtained during a drought year. Use a technique for estimating abundance that has untested assumptions. Avoid stating confidence limits. Under these conditions a coin toss would suggest that about half of the species should show declines. If yours does, publish. If it does not, find a post hoc explanation to account for the result. For example, hypothesize that dry weather reduced or increased activity and distorted your counts. Give your conclusion some statistical rigor by correlating weather phenomena with measures of abundance. But of course don't test the weather hypothesis against an independent data set.
Comments
Published in Conservation Biology. Volume 7, No. 4, December 1993. Permission to use.