Laserfiche WebLink
10 <br />COLORADO SCHOOL OF MINES RESEARCH INSTITUTE <br />the remaining strip production data to equal a fraction of the maximum produc- <br />tion. By scaling the data in this manner, response was unaffected by production <br />differences from plot to plot; thus all plots and all years contributed equally <br />to the average response pattern. <br />Standard statistical methods were used throughout this study. For much <br />of the data, the paired -t test was the most appropriate statistic for data <br />evaluation. In using this statistic, a pair consisted of one observation for <br />spoil and one observation for soil at the same plot; variation from one pair (or <br />plot) to another was of no interest (when the paired -t test was used) and was <br />eliminated by using the pairing procedure. In the context of this study, soil <br />and spoil can be justifiably placed in the same population. Unless otherwise <br />noted, the significance level was set at 0.05 <br />1 <br />r <br />1 <br />r <br />r <br />r <br />1 <br />1 <br />1 <br />