The sensitivity ratio: A superior method to compare plant and pathogen screening tests
There are numerous plant disease-screening methods used to identify resistance in various crops. It is common practice to prefer the screening method with the smallest root mean square error (RMSE), least-significant difference (LSD), or coefficient of variation (CV). However, valid comparison based on the RMSE or LSD requires both methods to have the same scale while the CV is only applicable if the scales are proportional to each other. Most plant disease-screening methods developed for the same disease have different scales and, generally, it is not clear that the scales are proportional. The sensitivity ratio is a statistic specifically developed for comparing different measuring methods and is not based on any particular assumption about how the scales are related. We use the sensitivity ratio to compare two white mold [Sclerotinia sclerotiorum (Lib.) deBary] resistance screening methods on soybean [Glycine max (L.) Merr.], and two methods of evaluating pathogen isolate aggressiveness on dry bean (Phaseolus vulgaris L.). Our results suggest that the sensitivity ratio should be used when comparing plant disease-screening methods.