Hypothesis Tests |
||
Main Concepts | Demonstration | Activity | Teaching Tips | Data Collection & Analysis | Practice Questions | Milestone | Fathom Tutorial | ||
Milestone When you get sick, you take your temperature. And what value do you compare this temperature to? Probably you use 98.6 degrees Farenheit, assuming that that is the "healthy" temperature. But body temperatures of healthy people vary from person to person (and they very within a person too -- your temperature in the middle of the night is different from in the middle of the day). In fact, this number should be interpreted as the mean body temperature of healthy people. But is it correct? The number comes from a study conducted in the 19th century by Carl Friedrich Wunderlich and is based on millions of measurements. (He measured a large number of people many times). At the end of his study, he pronounced 98.6 degrees as the mean temperature of healthy people. (Well, okay. He used the Celsius scale, and we translated it to Farenheit.) A much more recent study, published in the Journal of the American Medical Association, set out to try to confirm (or refute) this magic number. You can download the data yourself from http://inspire.stat.ucla.edu/unit_11/bodytemp.txt The file is a tab-delimited text file, and so you should be able to import this directly into Fathom using the "Import from URL" feature, simply by typing this URL into Fathom. 1) Does this data confirm that the mean body temperature is 98.6? In addition to doing the hypothesis test, state and check your assumptions as well as you can. Use graphs, or whatever means you feel appropriate. 2) Is there evidence that healthy body temperatures differ for men and women? We'll talk about how to do this formally (with hypothesis tests) in the next unit. For now, examine the data as best you can and state your belief, giving enough supporting evidence to be convincing.
NOTE: These data are actually not quite "real". After completing the milestone you might want to read about the data set in the Journal of Statistics Education. Don't read until you're done, since the article will spoil some of the fun. Write up your results and name them MS11yourlast name and drop in the Dropbox. |