HOOSING THE BEST MEASURE: Suppose you applied to be a waiter or waitress at a local restaurant and asked the manager what a typical dinner shift was like. How might the manager describe the typical size of a party at a table? The mode would be useful, since there are a few repeated values. You might have an occasional table for 20, but, for the most part, parties consist of one to five people. What if you asked for the typical size of a check, so you could estimate your tips? The mode makes no sense, since you would not likely have many repeats, and the mean is easily distorted by that table for 20, but the median tells you that half the time you can expect to earn a certain amount of money in tips.
Now suppose you were a real estate agent and you were asked by a client about the “typical” home in a subdivision. Being the astute agent you are, you have gathered the following information on each house in the subdivision: Price, square footage, numbers of bedrooms, number of bathrooms, and age. What statistic (mean, median, or mode) would you use to describe each aspect of the typical home and why? Try to imagine the type of answers you would be giving your client based on your selections for a subdivision that has 100 homes with a wide variety of sizes and prices.
Do you always trust the statistics you see in print? How about the ones you hear on the news or on your favorite talk show? What makes the statistics credible? For this question, I want you to tell me what personal criteria you would (or should) use to determine if you should believe reported statistics. Be specific, giving examples as needed.