Laserfiche WebLink
a <br /> TPH <br /> x <br /> a <br /> i <br /> O <br /> 0 <br /> -z o z <br /> Theoretical Quantiles(Standard Normal) <br /> vo,more rortnanoon th—Wot:.ao.wi coma I,,nada Qu ,t9l...t-1.EPA QNG 9.N,z.3.1 th—gh z.Yu.�l�.//wVlMl epa.gov/quality/ga-docs,htm1). <br /> Tests for TPH <br /> A goodness-of-fit test was performed to test whether the data set had been drawn from an underlying normal distribution. <br /> The Shapiro-Wilk (SW)test was used to test the null hypothesis that the data are normally distributed. The test was <br /> conducted at the 5% significance level, i.e., the probability the test incorrectly rejects the null hypothesis was set at 0.05. <br /> NORMAL DISTRIBUTION TEST <br /> Shapiro-Wlk Test Statistic 0 <br /> Shapiro-Wlk 5% Critical Value 0.866 <br /> The calculated SW test statistic is less than the 5% Shapiro-Wilk critical value, so we can reject the hypothesis that the <br /> data are normal, or in other words the data do not appear to follow a normal distribution at the 5% level of significance. <br /> The Q-Q plot displayed above should be used to further assess the normality of the data. <br /> Upper Confidence Limit on the True Mean <br /> Two methods were used to compute the upper confidence limit (UCL) on the mean. The first is a parametric method that <br /> assumes a normal distribution. The second is the Chebyshev method, which requires no distributional assumption. <br /> UCLs ON THE MEAN <br /> 95% Parametric UCL 0 <br />