PARENT SESSION
Poster Session #29: Animal Population and Community Ecology.
Wednesday, August 8, 2001. Presentation from 3:30 PM to 5:00 PM. Exhibition Hall

Understanding community structure with two new methods: transparent error and transparent observer.

Meyer, Eugene¹, ¹

ABSTRACT- Visitors to this interactive poster are invited to compare two methods of understanding a community of reptiles and amphibians. Visitors can place a transparency upon two graphs of community data. Visitors can add artificial errors on top of scatter graphs of real data that remain unchanged. The first graph represents an actual community in a Maryland marsh. The second graph represents experimental search results for an artificial community made for the purpose of measuring detectability. By writing on blank transparencies over the graphs, it is easy to see different categories of errors and how to correct them. The visitor can see how impacts vary by problem, including omissions, inflated values, and false zeroes. I tentatively name two problems, unreported in the literature, as column-swap and row-shift. These problems can nullify or even reverse a study's conclusions, and fortunately are easy to detect if you know how to look for them. A special case is any cell's error propagating to a wrong conclusion. Because the largest study of that problem's frequency occurs in a US Treasury study, I call it tax-return propagation. Readers are alo invited to superimpose transparencies of real data showing observer differences in 2D and 3D graphs. This transparent observer method shows when observer differences are positive, creating a measurably more accurate record of populations and communities. The goal is to understand more of the biology of these communities.

KEY WORDS: accuracy, camouflage, amphibian, rare