You were planning to run a one-factor ANOVA on the results of a clinical trial but the sample data for three out of your four treatments (compounds A, B, and D) exhibits positive skewness > 1.79 and the compound C response distribution is platykurtic, with excess kurtosis of -3.49. You run more tests and find Shapiro-Wilk scores (?=0.05) for the treatment group range between W=0.960 and W=0.982 with p-values ranging from 4.36e-3 to 0.201.

(a) Interpret the skewness, kurtosis, and Shapiro-Wilk test resultsand in 1 additional sentence explain why those results matter to your planned ANOVA.

(b) There are a number of options for dealing with a situation where necessary assumptions to runa valid statistical test are violated - for example, if the deviations are small enough, you can oftenignore them. In rare circumstances and after careful deliberation you may even be able to excisegenuine outliers to fix the problem. Describe one other option fordealing with data that does not satisfy the necessary assumptions to properly run a statistical testlike ANOVA.

Drug or control (C) size final drop % reduction A 32.67 21.19 11.48 35.13 A 23.67 22.76 0.91 3.85 7.69 6.7 0.99 12.86 22.51 19.8 2.71 12.02 32 29.43 2.57 8.03 19.01 15.66 3.35 17.6 13.53 10.64 2.89 21.35 6.61 6.12 0.49 7.37 22.66 14.83 7.83 34.57 8.46 4.51 3.95 46.7 23.45 18.57 4.88 20.81 33.97 25.17 8.8 25.91 21.58 15.21 6.37 29.5 nononnon ww ww ww ww > > > > > > 5.4 4.58 0.82 15.28 25.98 16.38 9.6 36.95 12.09 10.51 1.58 13.07 10.73 10.82 -0.09 -0.83 12.09 11.75 0.34 2.78 30.08 29.27 0.81 2.68 11.08 10.79 0.29 2.6 30.65 30.24 0.41 1.33 11.89 12.45 -0.56 -4.67 28.44 27.54 0.9 3.17 11.89 12.45 -0.56 -4.67