QUESTION 42 Questions 39 - 43 concern the following idea: A sheep farmer in WA is considering changing breeds of sheep in the hope they will produce more wool (call the two breeds "new" and "old"). 10 sheep of each variety are raised in the same paddock and their first fleece is weighed. The data is analysed in R via a 2-sample t-test with a 2-sided hypothesis, and the output is summarized below (with some parts obscured with ??). Two Sample t-test data: weight by type t = 1.9387, df = ??, p-value = 0.06838 alternative hypothesis: true difference in means is not equal to 0 95 percent confidence interval: ?? ?? mean sd n new 5.68 0.838 10 old 4.98 0.776 10 Levene's Test for Homogeneity of Variance (center = "mean") Df Pr(>F) group 1 0.0012 0.973 18 Let uA and up denote the mean amount of wool produced from the "new" and "old" breeds of sheep respectively, and OA and Op be the respective standard deviations. In answering the questions below, use a = 0.05 where necessary. Bob thinks it would be a good idea to extend this trial and get a second fleece from each sheep, and then analyze the data in the same way with 20 fleeces for each sheep breed. Why is Bob wrong? O a. Bob is wrong as getting two fleeces from each sheep would increase the degrees of freedom which would make the data more variable and so more likely to violate the assumption of constant variance. O b. Bob is wrong as getting two fleeces from each sheep would likely violate the assumption of normality. O c. Bob is wrong as getting two fleeces from each sheep would still leave the sample size