For problems 7-9, consider the StatKey dataset CocaineTreatment from the 3rd edition of the Lock text. 7. 9. If we assume the subjects were volunteers who were randomly assigned to the treatments. i. Is the design of this study appropriate to support generalization to the population of cocaine users? a. Yes b. No ii. Is the design of this study appropriate to support evidence for causality if a sufficiently strong effect is found? a. Yes b. No Test the claim that the proportion of people in the population treated with lithium have a higher proportion of non-relapse than those who are not treated (i.e. treated with a placebo in our sample.) Find the p-value for this test. a. 0.27 b. 0.36 c. 0.54 d. 0.65 e. 0.72 Find a 83% confidence interval for the difference in the proportion of people in the population who are expected to have no relapse between those treated with desipramine and those treated with lithium. Then subtract the two endpoints of the interval to report the length of the interval here. a. 0.21 b. 0.35 c. 0.44 CI. 0.54