Use of Power Curves to Monitor Range Trend

Abstract

Type I and Type II errors and the power of the test when testing the null hypothesis of static range trend are discussed. The consequences associated with Type I and Type II errors are judged to be similar and therefore the probability of committing a Type I or Type II error should be equal. As an example, the current range trend monitoring program for the Moose Camp Allotment on USDA Forest Service land in southwestern Montana is capable of detecting a change in range condition of one condition class 83% of the time when the probability of Type I error is set at .17 (Prob[-Type I Error] = Prob [Type II Error] = .17). Doubling the sample size would increase the ability to detect a condition class change to 95% when the probability of Type I error is set at .05 (Prob[Type I Error] = Prob[Type II Error] = .05).