In-Class Group Project Part 1



In-Class Group Project Part 2

(due at the end of class)

The in-class data analysis project will be examining the relationship between race and prevalent diabetes. The data for this project come from the 2007 Behavior Risk Factor Surveillance System (BRFSS) located on the CDC website for the states of South Carolina and Pennsylvania. A subset of the BRFSS dataset has been place on the class website.

Project part 2 relates to becoming familiar with the exposure and outcome variables. This exercise will help you decide how to categorize your variables. One thing to consider when categorizing variables is missing data. Prior to class you should have read Chapter 4, “Planning the Measurements: Precision and Accuracy.”

Defining Diabetes – The Outcome Variable:

1. What variables are there in the BRFSS that you might use to define diabetes?

2. What percentage of individuals was classified as having diabetes only when pregnant?

3. Were any men classified as having diabetes while pregnant?

4. What is the prevalence of borderline diabetes?

5. Should individuals with diabetes only during pregnancy or individuals with borderline diabetes be included as individuals with diabetes for analysis purposes? Justify your answer.

6. How many individuals were asked if they were taking pills to treat their diabetes? Who were these individuals?

7. How many people of those asked if they were taking pills to treat their diabetes were missing information on whether or not they took pills to treat their diabetes? Please itemize the different types of missing.

8. Are the people missing information on whether or not they took pills to treat their diabetes the same people that are missing information on whether or not they took insulin to treat their diabetes?

9. Create a variable that indicates whether or not an individual was taking pills and/or insulin to treat their diabetes? How will you classify people missing information from one or both of these variables?

10. How many individuals were asked their age of diabetes onset? Who were these individuals?

11. How many people of those asked their age of diabetes onset were missing information on age of diabetes onset? Please itemize the different types of missing. What is the oldest age of diabetes onset recorded in the dataset?

12. Among people who report having diabetes what percent are missing age of diabetes onset? Among people who report having diabetes are people who report age of onset different from those who do not? Choose three characteristics and evaluate this.

Defining Race/Ethnicity – The Exposure Variable:

13. What questions are there in the BRFSS that you might use to define racial/ethnic groups? What does the raw output look like for these questions? Why do only a few people (about 1%) report the race that best represents them?

14. For variable _RaceGR2 define each category. What does ‘multirace’ imply?

15. Describe the racial/ethnic characteristics of people in SC and PA using variable _RaceGR2.

16. For analysis purposes how do you recommend defining race/ethnic group? Think about what the focus is of the project and use variable _RaceGR2 as the starting point (not the raw data)? Justify your answer.

................
................

In order to avoid copyright disputes, this page is only a partial summary.

Google Online Preview   Download