Abstract

When there are several methods of measuring a physical or chemical property, it is necessary to determine which method is best. If both methods are measured on the same scale, the most precise method will be preferred. However, often the methods have different scales. The sensitivity ratio allows for explicit comparison of methods with different scales. We use the sensitivity ratio to compare soybean resistance screening methods to evaluate the resistance of soybean varieties to Sclerotinia sclerotiorum. When compared to the root mean square error or the coefficient of variation, the sensitivity ratio can order methods differently both when the methods are measured on different scales and on the same scale. Our results cast doubt upon using standard precision statistics such as the root mean square error or the coefficient of variation to compare measuring methods and we suggest that the sensitivity ratio should be used instead.

Creative Commons License


This work is licensed under a Creative Commons Attribution-Noncommercial-No Derivative Works 4.0 License.

Share

COinS
 
Apr 27th, 2:30 PM

COMPARING MEASURING METHODS USING THE SENSITIVITY RATIO: AN APPLICATION TO RESISTANCE SCREENING IN SOYBEANS

When there are several methods of measuring a physical or chemical property, it is necessary to determine which method is best. If both methods are measured on the same scale, the most precise method will be preferred. However, often the methods have different scales. The sensitivity ratio allows for explicit comparison of methods with different scales. We use the sensitivity ratio to compare soybean resistance screening methods to evaluate the resistance of soybean varieties to Sclerotinia sclerotiorum. When compared to the root mean square error or the coefficient of variation, the sensitivity ratio can order methods differently both when the methods are measured on different scales and on the same scale. Our results cast doubt upon using standard precision statistics such as the root mean square error or the coefficient of variation to compare measuring methods and we suggest that the sensitivity ratio should be used instead.