Use LEFT and RIGHT arrow keys to navigate between flashcards;
Use UP and DOWN arrow keys to flip the card;
H to show hint;
A reads text to speech;
9 Cards in this Set
- Front
- Back
|
What are "pre-market" factors? |
Productivity differences: essentially means that the discrimination occurs earlier in life than when wages are set |
|
|
What is Statistical discrimination? |
Employers perceive productivity differences across groups |
|
|
What is Preference discrimination? |
Employers don’t like hiring from other groups |
|
|
What makes the gender wage gap different from the racial wage gap? |
-One particularly important factor: it cuts across socioeconomic status
-Pre-market factors come from a very different place and play a different role
More comparable situations internationally |
|
|
What is the interpretation of b here:
Earnings = a + b*Male? + c*Education + e |
b here represents the average male wage premium for workers with the same level of education. |
|
|
Given the regression: Earnings = a + b*Male? + c*Education + e
What would it mean if b in this regression were smaller than in the previous regression? Larger? |
Smaller: than overall, it means that some of the overall wage gap is explained by the fact that men are pursuing higher levels of education.
Larger= it must mean that the gap is even worse than it looks because women are pursuing higher levels of education. |
|
|
Given the regression: Earnings = a + b*Male? + c*Education + e
In the real world, do you expect b to be larger or smaller here than in the previous regression? |
This could go both ways: women are now the majority of undergraduates, but many professional degrees are still majority male. |
|
|
Given the regression: Earnings = a + b*Male? + c*Education + d*Hours + e
Is a b > 0 indicative of explicit discrimination? |
No.
Holding hours constant in a linear fashion may miss the fact that per-hour wages increase with hours worked. This may represent explicit discrimination, but it may also represent long-hours bonuses and premiums |
|
|
Remembering what we learned about wages for working long hours, how might you adjust the regression:
Earnings = a + b*Male? + c*Education + d*Hours + e to do a better job looking for explicit discrimination? |
To adjust for the fact that in many occupations, hourly wages increase with hours work, you could add the variable hours^2 to the regression:
Earnings = a + b*Male? + c*Education + d*Hours + f*Hours^2 + e
This allows for two different coefficients on hours. One represents the “linear” increase in earnings as hours go up.
The other represents the “non-linear” increase in the hourly wage as hours go up. |
