Fall Semester 2004

Chapter 10: Analysis of Variance.

1. (Fill in the blank). From the book: "As you will see [in this chapter] analysis of variance can be thought of as a special case of ___________".

2. Some professional associations are reluctant to hold meetings in New York City because of high hotel prices and taxes. Are hotels in New York City more expensive than hotels in other major cities? To answer this question, let's look at hotel prices in four major cities: Los Angeles, New York City, San Francisco, and Washington DC. For each city, a random sample of eight hotels was taken from the 1992 Mobil Travel Guide to Major Cities and stored in the workbook Hotel.xls. Describe how you would use this data to give an answer to the original question (about hotels in NY being more expensive than in other major cities). Be succinct but also very clear (not ambiguous).

3. Is there anything special about San Francisco in the data set? What do we need to do about it (if anything)?

4. About how much of the variability in hotel price is explained by the city of origin?

5. In performing the analysis do you need to calculate pairwise mean differences? Why or why not? What is the Bonferroni correction factor and why (and when) would you need to use it? What alternative methods of comparing means do you have? What is their benefit (if any)?

6. Describe the problem and the solution for the two-way analysis example used in the book. Is there or isn't there significant interaction between cola type and cola brand in the example? What's the reason for your conclusion.