Sunday, September 11, 2005

-- Science and Sample Size --

----------------------------------------------------------------------------------------------------------
One of the fundamental tenets of science methodology concerns having adequate sample sizes from which to draw conclusions/generalizations. In the years since James Tanner's dissertation on the Ivory-bill (based on but a handful of birds), notions that Tanner himself often recognized as tentative became hardened into unchallenged dictums without a good basis for doing so. There is in fact little that can be stated with certainty about the Ivory-bill's diet, behavior, habits, or requirements for survival, even though such statements are rife in the literature. (If one were to intensely study a dozen people and then write a report generalizing to the entire human species the weakness would be readily apparent.) This is all especially true given that any Ivory-bills still around today may in fact have survived specifically BECAUSE they came from individuals with significantly DIFFERENT characteristics/behavioral traits from their brethren, which increased survivability for themselves and their offspring. At least Tanner got it right at the end of his original introduction:
"The chief difficulty of the study has been that of drawing conclusions from relatively few observations... My own observations of the birds have been entirely confined to a few individuals in one part of Louisiana... the conclusions drawn from them will not necessarily apply to the species as it once was nor to individuals living in other areas. The difficulty of finding the birds, even when their whereabouts was known, also limited the number of observations. Especially was this true in the non-breeding season. With these considerations in mind, one must draw conclusions carefully and with reservations." (italics added)

The problem with our knowledge of Ivory-bills is not simply how little we know, but rather how much we think we know that might just be utterly wrong for any birds remaining today...
---------------------------------------------------------------------------------------------------------

No comments: