From the assignment that I submitted, I was asked this question below directly. Can you provide an answer to this question by 8:00 a.m. on 8/12/16?
Great job in your Discussion. When you stated a standard curve is possible for for a probable conclusion, would you recommend an interpolation using a linear or nonlinear regression analysis? How did you decide to use the Kruskal-Wallis analysis? The Kruskal-Wallis is a non-parametric test, according to McDonald (2014), this test does not assume the data is normally distributed. McDonald (2014) argues if you use this to test if the medians are different, if different groups have different shapes or different variances, theKruskal-Wallis test may give inaccurate results. According to McDonald (2014), a Welch’st-test for two groups or a one-way anova is easier to comprehend and a lot more powerful.
Reference
McDonald, J.H. 2014. Handbook of Biological Statistics (3rd ed.). Baltimore, : Sparky House Publishing.
