Module 5: Multivariate displays
(The attached PDF file has better formatting.)
Homework assignment: conditioning plots
This data set has 7,425 observations and 4 variables: age, sex, education, and wages.
The dummy variable for sex has values 1 = female and 2 = male.
The ages range from 16 to 95.
The education is years of schooling (from 0 to 20)
The vertical axis is the logarithm of wages (2,300 to 49,920 Canadian dollars).
Each graph has two lines.
The straight (diagonal) line is a simple regression line.
The curved line is a lowess curve.
A. What is the dependent variable? What is the independent variable? What are the conditioning variables?
B. How much do the ages overlap? Note that upper bound of one age is the midpoint of the next age and the lower bound of the next age.
C. The graph has 12 panes. Explain what the graph in the top row and second column from the left shows. (A one sentence explanation is sufficient.)
D. The lowess curves match the regression lines at higher education levels but lie above the regression lines at lower education levels. What does this imply?
E. What group has a stronger relation of wages to education level: women age 38 to 54 or women age 23 to 38?
F. What group has a stronger relation of wages to education level: women age 38 to 54 or men age 38 to 54?