In 1936 Sir Ronald Fisher presented data on irises as the example in a famous statistics paper. Ever since, “Fisher’s Iris data” have been a feature of statistics texts. We didn’t want to be an exception. Can measurements of the petal length of flowers be of value when you need to determine the species of iris? Here are the summary statistics from measurements of the petals of two species of irises. (R. A. Fisher, “The Use of Multiple Measurements in Axonomic Problems,” Annals of Eugenics 7 [1936]:179–188)
a) Make parallel boxplots of petal lengths for the two species. b) Describe the differences seen in the boxplots. c) Write a 95% confidence interval for this difference. d) Explain what your interval means. e) Based on your confidence interval, is there evidence of a difference in petal length? Explain. *f) By resampling the data, create a 95% bootstrap confidence interval and compare your results to those above. Did your conclusion in part e change?
