These are the goals for week 12:
We will be working in chapter 9 on two sample tests. The main thrust of this chapter is to compare two different samples to see if there is a difference.
We will do a few more examples of hypothesis testing for one sample on Monday, and then begin doing examples of two sample tests. There will be both confidence intervals and hypothesis testing questions.
Please try to bring an article on comparison of data for Wednesday’s lecture. For example, on http://www.cnn.com I found an article about the relationship between moderate drinking and heart benefits. The data is presented as
In one of the studies, Dr. Kenneth J. Mukamal of Boston’s Beth Israel Deaconess Medical Center and colleagues interviewed 1,913 heart attack patients hospitalized nationwide between 1989 and 1994. The men and women were asked about their drinking habits in the preceding year. Death records through 1995 showed 317 died, about three-fourths from heart-related causes.
Compared with abstainers, light drinkers (under seven drinks a week) were about 20 percent less likely to die and moderate drinkers (seven or more drinks a week) were about 30 percent less likely to die.
It made little difference whether the patients drank mostly wine, beer or liquor.
The researchers said the numbers were too small to evaluate the effects of heavy drinking.
We could make this a two sample test, by comparing the abstainers to the drinkers. (Otherwise, it is a bit more complicated involving ANOVA).
There are quite a few homeworks for the week. Many of the computational ones will be done in lab.
9.5, 9.7, 9.8, 9.13, 9.22, 9.25, 9.29, 9.38, 9.40, 9.49, 9.50, 9.62 ($s_d=2.17$)