Location: Fruit and Tree Nut ResearchTitle: Use of the disease severity index for null hypothesis testing
|CHIANG, KUO-SZU - National Chung-Hsing University|
Submitted to: American Phytopathological Society Annual Meeting
Publication Type: Abstract Only
Publication Acceptance Date: 5/17/2017
Publication Date: 12/1/2017
Citation: Chiang, K., Bock, C.H. 2017. Use of the disease severity index for null hypothesis testing. American Phytopathological Society Annual Meeting. 107:S5.6.
Technical Abstract: A disease severity index (DSI) is a single number for summarizing a large amount of disease severity information. It is used to indicate relative resistance of cultivars, to relate disease severity to yield loss, or to compare treatments. The DSI has most often been based on a special type of ordinal scale data defining a series of consecutive numeric ranges, generally based on the percent scale. The objective of this work is to explore the effects both of different ordinal scales (those having equal or unequal interval widths) and of the selection of values for scale intervals (the ordinal score for the interval range or the midpoint value) on the null hypothesis test for treatment comparisons. A simulation approach was employed to approximate the mechanisms governing the estimation process. Subsequently, a meta-analysis was used to compare two treatments, demonstrating that using quantitative ordinal rating scores or the midpoint values of interval yielded very comparable results with respect to the power of hypothesis testing. The main factor determining the power of the hypothesis test is the nature of the interval widths (equal or unequal), not the selection of values for ordinal scale intervals (score or midpoint value). Although the percent scale is preferable, the results provide a framework for developing improved research methods where the use of ordinal scales in conjunction with a DSI is preferred for comparing disease severities.