VIEW ARTICLE

Letter to the Editor

Two-Dimensional Distance Class Analysis of Disease-Incidence Data: Problems and Possible Solutions. F. J. Ferrandino, the Connecticut Agricultural Experiment Station, New Haven 06504. Accepted for Publication 18 April 1996. Phytopathology 86:685-691. Copyright 1996 The American Phytopathological Society. doi:10.1094/Phyto-86-685.

Many recent papers on disease incidence have used a maplike representation of deviations from random behavior. The basic procedure is to compare observed infected pair counts with expected pair counts, assuming a random distribution. For the large number of multiple comparisons made in these analyses, the level of confidence used is not appropriate. This letter discusses the problem and proposes a new method of analysis, along with directions for its use.