Due: November 7 at noon turn in the .html file on Sakai
For this homework, use the MedGPA dataset in the Stat2Data package.
You can find out more about this dataset by running ?MedGPA in R after loading it.
Describe the data. What are the observations? What are the variables?
Using the quantile() function, what are the 0.2, 0.4, 0.6, and 0.8 quantiles?
You can find out more about the case_when() function by running ?case_when in R. Make sure you have loaded the tidyverse package.
case_when() and mutate() functions, create a new variable called gpa_cat that categorizes GPA into 5 categories. Use the quantiles above to define your categories. For example, if you had a variable var with the quantiles:You would create the categories like this:
new_cat = case_when(
var < 5 ~ "quantile 1",
var >= 5 & var < 6 ~ "quantile 2",
var >= 6 & var < 8 ~ "quantile 3",
var >= 8 & var < 11 ~ "quantile 4",
var >= 11 ~ "quantile 5"
)Calculate the average GPA in each category.
Calculate the proportion where Accept == "A" for each of the categories created in Exercise 3.
One of these categories has 100% Acceptance. This makes it hard to calculate the odds. Let’s add a “fudge factor”. Calculate “adjusted proportions” using the following equation.
\(p = \frac{(1/2 + \#Accept = A)}{(1 + \#Accept = A + \#Accept = D)}\)
Using the adjusted proportions in Exercise 6, calculate the odds of Accept == "A" for each of the categories created in Exercise 3.
Using the odds calculated in Exercise 7, calculate the log odds of Accept = "A" for each of the categories created in Exercise 3.
Create a plot with mean GPA calculated in Exercise 4 on the x-axis and the log odds calculated in Exercise 8 on the y-axis. Does the linearity assumption of the log odds hold?
Create a plot with mean GPA calculated in Exercise 4 on the x-axis and the adjusted proportion calculated in Exercise 6 on the y-axis. Describe the shape that you see.
Run a logistic regression model estimating Accept == "A" using GPA. Interpret the coefficient for GPA.
Rank each of the following principles from the Elements and Principles of Data Analysis article for this data analysis from 1 to 10 along with a one sentence summary: