STATISTICAL TREATMENT OF SOIL CHEMICAL CONCENTRATION DATA

Soil chemical field data typically do not satisfy the required statist ical assumptions, and this renders statistical tests based on normal t heory either invalid or not particularly powerful. The objective of th is study was to compare the t-test and two nonparametric tests (Wilcox on signed rank and the Sign test) for a theoretical data set and 3 yr of soil atrazine 2-chloro-4-ethylamino-6-isopropylamino-s-triazine) co ncentration field data, to demonstrate how the sample distribution aff ects the statistical analysis. The theoretical data set was contructed to emulate a soil chemical data set in which a minimum detection limi t resulted in multiple zeros within the data. These data were non-norm al, and the normal-theory tests were not valid. The converse was also demonstrated. The performance of the nonparametric tests was evaluated when the data were from a normal distribution. The Wilcoxon signed ra nk test performed well on normal data, although there were some data f or which the t-test was more powerful. Actual soil atrazine concentrat ion data from 78 sampling events were analyzed both with the paired t- test, Sign test, and Wilcoxon signed rank test, Thirty-three percent o f the events were not from normal distributions, and 15% of these resu lted in different decisions regarding the null hypothesis if the paire d t-test was used instead of the Wilcoxon signed rank test. Of the 66% of data sets that were from normal distributions, 5.7% of these resul ted in different decisions regarding the null hypothesis if the Wilcox on signed rank test was used instead of the t-test. It is recommended that all soil chemical data sets be tested for normality. If the data are not normally distributed, the appropriate nonparametric test shoul d be used rather than attempting to transform the data to normality. S everal other nonparametric tests are presented.