(1) Testing for streaks or slumps. (2) Correlation using Data Desk. (3) How the poll was conducted. (4) Interview with Judith Tanur.
Today is election day and the New York Times has summarized the recent polls for the Governor's race in New Jersey and the race for Mayor in New York . See attached sheet.
1. The ``plus or minus 4 percentage points" is more normally stated as ``This poll has a margin of error of plus or minus 4 points." What does this mean?
2. Can all these polls be correct? Will we know which were correct after the election?
3. Note that these polls are based on the opinions of about1000 likely voters. How do you think these 1000 people were chosen?
4. Yesterday's Valley News stated that ``Last year, a national survey showed that 56 percent of Dartmouth students drink at least once a week, compared to 45 percent of students nationally. Ten percent of Dartmouth students reported averaging 21 or more drinks a week, compared to 5.8 percent nationally." For your project, you decide to conduct your own survey to see if these percentages are accurate for Dartmouth students. How would you choose your sample?
The Chance fair will be held on Sunday December 5th from 4:00 to 6:00 in Alumni Hall in Hopkins Center. The main purpose of this fair is to show off the work that you did on your final project though you will also be given a chance to exhibit your skill at certain games of chance.
For this Chance fair you should prepare a poster that will explain to us and your fellow students the principal results of your study. Your poster should be enhanced by the use of bright colors, pictures, graphs etc., and may be accompanied by computer displays, movies or anything else that might be appropriate to help explain your project.
You might be able to report all the results of your research on your poster but, more typically, you will want to prepare a separate write-up to give the results of your investigation in more detail. You can attach a copy of this paper to your poster for those at the fair who want to see more detail but be sure to also turn in to us a copy of this supplementary write-up.
You may work with another student or two on your project but each person should have primary responsibility for a significant part of the work and your write-up should explain this division of responsibility.
We will exhibit in class a few of last year's posters.
As we stated at the beginning of the course, your project may be a paper investigating more deeply some topic we touch on lightly in class, or a topic that might arise in one of your other classes or any topic of special interest to you that involves chance. Alternatively, you could design and carry out your own study. Or you might choose to do a computer-based project. If you do your own study a simple description of the results will not be sufficient. You must include some statistical analysis of the results. We can help you with this.
Suppose you have a histogram or some other display from Data Desk and you want to print. The easiest way to do this is to select the object you want to print by just clicking anywhere on the window containing it. The upper bar of the window should be dark after you have done this. If you have just made it, the upper bar is probably is already dark and you do not have to select it. You then choose ``Print window" from the File menu. If you want to put your result in a Word document you can, instead, choose ``Copy window" from the file menu and paste it in the Word document.