Question 11 0.5 pts When conducting hypothesis tests involving a multilevel categorical variable, we often use the chi- squared (X2) test rather than conducting pairwise comparisons between every possible pair of levels of that variable using Z tests. Why? (Select all that apply) C] The risk of committing Type II errors increases for each null hypothesis we fail to reject C] The risk of committing Type | errors increases for each null hypothesis we reject C] The multiple testing problem increases quadratically as the number of levels increases C] Achi-squared test is more powerful than a Z test. Question 12 1 pts When we make statistical inferences about the expected value of numerical variables, IE[X], we often use the Student's T distribution rather than the Z distribution as a model of the standardized sample average even if we assume that the sample average is normally distributed around the expected value. Why? 0 There is no difference between the Z and the Student's T distribution; these are different names for the same distribution. 0 When we standardize the sample average, this is based on an estimate of the standard error based on the sample standard deviation rather than the population standard deviation. 0 While the 2 distribution is appropriate for unbounded numerical variables, only the Student's T distribution is appropriate for strictly positive numerical variables. 0 While the 2 distribution is appropriate for continuous numerical variables, only the Student's T distribution is appropriate for discrete numerical variables. Question 13 1 pts When we use the Student's T distribution to calculate a 95% confidence interval for the expected value of a numerical variable, how does sample size influence the T score for the 95% confidence level? O As sample size increases, the T score decreases toward 1.96. O Sample size does not influence the T score; it always equal 1.96. O As sample size increases, the T score increases toward 1.96