Here are some sample test questions for Chapter 10.

1. If the margin of error in an estimate for the mean weight of a shipment is + or - 2 pounds at a confidence level of 95 percent, what will be the margin of error at a confidence level of 98 percent?

2. Suppose that in a survey we find that 37 percent of people prefer Shaul's chicken to Shalom's. How large a sample would you need to know that the margin of error for a 95 percent confidence interval is no more than 3 percent? If the confidence level were 90 percent, would the minimum sample size be larger or smaller?(Hint--you can calculate the standard deviation because it is a binomial model.)

3. A school claims that 80 percent of its students get into one of their top three choices for college. The statistics teacher decides to test this at a 5 percent level of significance. Comment on the validity of each of the following interpretations of the 5 percent significance level:

   • 5% is the probability of a Type I error.
   • 5% is the probability of rejecting the hypothesis that the probability of getting into one of the top three choices is 0.8 when that hypothesis is true
   • 5% is the probability of accepting the hypothesis that the probability of getting into one of the top three choices is 0.8 when that hypothesis is false
   • 5% is the criteria for judging whether the sample statistic is sufficiently different from .8 to cast doubt on the school's claim

4. The null hypothesis is that the average high temperature in December is 45 degrees. If someone reports that the average was 52 degrees and that the P-value for this is .036, what does the P-value represent?

5. Will the confidence interval always be symmetric around the true population parameter? Explain.