Question 5:

Health economists are interested in understanding the cost for heart valve replacement at hospitals in Washington, DC. They want to test the hypothesis that the mean cost for heart valve replacement is greater than $150, 000. There is a debate on what significance level to use to conduct the hypothesis test, either 6% or 2%. Data fare collected from a sample of 100 patients who had heart valve replacement surgery in Washington, DC in the last year. There is failure to reject the null hypothesis at the 6% significance level, concluding that there is insufficient evidence in the data to claim that the mean cost for heart valve replacement is greater than $150, 000. Using the same data and statistical test, what will the conclusion be at the 2% level of significance? a. Fail to reject Ho b. Reject Ho c. Fail to reject Hi given Ho true d. Fail to reject Hi given H| true

e. There is not enough information given to determine the result

Question 6:

I the null hypothesis is rejected at the 6% level of significance in Question 5, which one of the following statements are true? a. Only a Type I error is possible b. Only a Type II error is possible c. Both Type I and Type Il errors are possible d. Neither a Type I or Type Il error is possible e. You can be 100% sure that you made the correct decision

Question 7:

Which one of the following statements are true with respect to the central limit theorem? a. The sampling distribution of the sample mean is only normally distributed if the sample mean is based on a small sample size b. The sampling distribution of the sample mean is only binomially distributed if the sample mean is based on a small sample size c. The variance of the sampling distribution of the sample mean will always be larger than the variance of the individual population values d. If the individual population values are normally distributed then the sampling distribution of the sample mean will be skewed to the right e. If the individual population values are not normally distributed then the sampling distribution of the sample mean will be normal if the sample mean is based on a large sample size