1. A company is interested in estimating the mean amount of money employees spend on lunch in their cafeteria. From a previous study we know that the population standard deviation is calculated to be $1.47. How many workers should be sampled so that a 90% confidence interval for the average cost of lunch at the cafeteria will have a margin error of $0.50 ?

2. National data in the 1960s showed that 48% of the adult population had never smoked. In 2008 a national health survey interviewed a random sample of 885 adults and found that 50.7% had never been smokers. Does this provide evidence of change in behaior among adults?

a) State the null and alternative hypotheses: Ho= Ha=

b) Conduct a test of the hypotheses identified above.

1. Calculate the test statistic.

2. Corresponding p-value.

3. Make and justify a statistical decision for the level of significance alpha=0.1

4. Interpret your decision in the context of the problem.

5. If your decision above was a mistake, what type of error would you have made?