Due Dates:

· Monday 11/24, in class - Brief Description of Proposal (10 pts)

· Friday 12/5, in class - Written Report (10 pts) and Brief Oral Report (10 pts)

Purpose: To answer a question of interest to you using one of the following statistical inference methods learned in this class:

· difference of two independent means

· mean of matched pairs differences

· difference of two independent proportions

Instructions:

1. You MUST work in groups of two or three people.

2. Decide on the question you will attempt to answer, and the two groups that you will compare. Here are some examples, just to give you some ideas:

Compare the average number of hours per week spent exercising by male and female UF students (two independent means).
Compare the average number of credit hours per semester taken by UF students who participate in the Honors program, and those who do not (two independent means).
Compare the average number of alcoholic drinks per week for UF students who belong to a fraternity/sorority and those who do not (two independent means).
Compare the proportion of people who prefer diet soda over regular soda for those under 21 and over 21 years old (two independent proportions).
Compare the proportion of white people and black people who voted for Barack Obama in the 2008 presidential election (two independent proportions).
Compare the average prices at two stores (or Internet merchants) based on a random sample of identical items from each one (matched pairs mean).
Compare the average time to complete a task using two different methods (in random order), or before and after some training/practice (matched pairs mean).
Compare the number of bicep curls that people can do using a 10 lb dumbbell with their dominant vs non-dominant arm (matched pairs mean).

3. Carefully design a survey or experiment to collect the data.

Pay particular attention to the selection of the random sample, avoiding possible sources of bias.
If you are conducting an opinion survey, be specific about where and how the data will be collected, and by whom, and what will you do about nonresponse.
If you are conducting an observational study or an experiment, be specific about how your measurements will be taken and recorded accurately.

4. Present your proposal to the class MONDAY 11/24. This will be an informal, five-minute presentation, worth 10 points.

5. Collect the data, making sure you have enough observations to satisfy the assumptions. For a problem about means, you should have 30 or more observations (so you don’t need to worry about normality). For proportions, you should have at least 15 successes and 15 failures in each sample.

6. Analyze the data, using Minitab. (Minitab is available in all the CIRCA labs that have PC’s, and can also be downloaded for free, for 30 days, to your computer. See course webpage for more information). When using Minitab, make sure you know which type of analysis is appropriate for the data you collected. You should make both a confidence interval and a significance test, and any relevant graphs and numerical summaries of your data. Note: You can easily copy the graphs, data and results from Minitab into a word processor using the cut and paste functions provided by the programs. You can then email yourself the document, and work on the rest at home if you want.

7. Friday, December 5^th:

Short ORAL presentation of your results to the class (10 points)

Turn in a WRITTEN report in class (10 points).

The report should include:

The names of all the students in the group.
A statement of the purpose of the project – why did you choose that question, and what did you expect to prove.
The data collected.
Relevant graphs that describe your data set.
The assumptions necessary for the validity of your test, and a statement of whether they seem to be satisfied or not.
The statistical analysis performed in Minitab.
Your conclusions (both statistical, and in plain English).