Question

Problem 8: The Framingham Heart Study was a longitudinal cohort study of 5000+ men and women. One outcome of interest was fasting glucose levels. Glucose levels were categorized into three different categories:

Glucose Levels

     -Diabetes (glucose >126),

    -Impaired Fasting Glucose (glucose 100-125),

    -Normal Glucose

Several possible risk factors were also recorded:

Risk Factors

     -Sex

     -Age

     -BMI (normal weight, overweight, obese)

     -Genetics

To determine if each possible risk factor is related to glucose levels,researchers need to use an appropriate hypothesis test.

Test Choices

1. ANOVA

2. Chi-Square GOF

3. Chi-Square test for independence

4. Test for equality of means

5. Test for equality of proportions

6. Other

a. What test would be used to assess whether the different sexes(male and female) have the same proportions of the different glucose levels?

b. What test would be used to assess whether the different glucose levels have the same mean age?

c. What test would be used to assess whether the different categories of BMI have the same proportions of the different glucose levels?

Answer 1

The Chi-Square test of independence is used to determine if there is a significant relationship between two nominal (categorical) variables.

The one-way analysis of variance (ANOVA) is used to determine whether there are any statistically significant differences between the means of three or more independent (unrelated) groups.

a. What test would be used to assess whether the different sexes(male and female) have the same proportions of the different glucose levels?

3. Chi-Square test for independence

b. What test would be used to assess whether the different glucose levels have the same mean age?

1. ANOVA

c. What test would be used to assess whether the different categories of BMI have the same proportions of the different glucose levels?

3. Chi-Square test for independence