Empirical Analysis and Research Methods

DUE DATE: Tuesday, April 24

General comments about writing the assignment:

• This is an example of a formal writing exercise where you will prepare a formal report.
• When you are writing your reports, you should assume that the reader does not know what the guiding question is. In other words, each of your answers should be able to be read as a stand alone report. In addition, the report should be written for a general audience with a limited knowledge of nutrition and economics.
• All answers must be prepared and posted on your website.
• If you are including tables, graphs, diagrams, figures, each item should include a number, title, and a source.
• Here are some guidelines to writing a good report: https://www.wikihow.com/Write-a-Report. Since the topic is already assigned to you, you can skim Part 1: Selecting your topic and focus on Part 2: Researching your topic, Part 3: Prewriting your report, Part 4: Writing your report, and Part 5: Finalizing your report.
• Please use the free writing resources posted on this webpage to help you with your writing.

For all of the questions below you will need to use excel file titled “Statistics Example” that is available on the website.

1. First describe the data - what are the variables and the descriptive statistics?

2. Do hypothesis testing comparing the two means on the following:

• BMI differs by sex
• BMI differs between American and International students
• BMI differs between those that drink coffee and those that do not
• Exercise level differs by sex
• Exercise level differs between American and International students
• Exercise level differs between those that drink coffee and those that do not

Your answer should include clearly state hypotheses, calculations, and conclusions.

3. Perform simple Chi2 tests

• Drinking coffee and race
• Drinking coffee and year in college

4. Prepare a scatter plot between BMI and Exercise that includes a trend line. Is there a correlation between the two variables? Explain.

5. Run the following regression

• BMI = a + b1Exercise + d1Sex + d2International + d3Coffee

Report your results and explain the meaning of coefficients

• If you decide that a modified version of the above regression is needed explain why and run the appropriate new regression. Report your results and explain the meaning of coefficients.
• What are some of the problems with this regression in general? What could be done to ameliorate those problems?

Below is the data that you will need for this problem set.

In order to save it on your computer and use it in excel, please open the file, the click on File => Download As => Microsoft Excel (.xlsx) and save it on your computer.