Total Points: 8
- You are a researcher interested in the relationship between reading and income. You conduct a survey of young professionals (25-30 years old) where you obtain the average amount of time they spend reading outside of their job, and their annual income. You obtain the following regression where reading is in hours per week, and annual income in thousands of dollars;
Select the arguments that would suggest the estimated effect of reading on income is not causal or has some bias (Select all that apply.)
A. That those who read more outside of their job also prioritize other hobbies and activities, a prioritization which on its own reduces their income.
B.That those who read more are more likely to be in a Barnes and Noble, which would increase the number of books they own.
C. That those who read more are also more motivated and self-disciplined, traits which on their own increase income. D. That people with more connections are going to be getting better paying jobs, which affects income on its own.
E. People who have higher income, tend to buy more books and therefore read more often, as opposed to readings changing one’s income.
A, C, and E
- Why might there be differences between individuals’ actual outcomes Y, and their predicted outcomes Y from a regression? (Select all that apply):
A) There is random “noise” or error in sampling and measuring individual outcomes
B) Our sample is too small too be representative of the population
C) The observations that do not fall on the line of best fit are outliers
D) The differences are due to a bad fit of the data, if we had another line we would not have these errors.
A
- 3) Which is true of the following conditional expectation expression? Assume both variables are binary. (Select all that apply):
- There is a promising new white noise machine that has been shown to help increase the hours people sleep. The developers of this product want to grow sales, but an investor is concerned that people who bought this machine may also be more likely to use other sleep aids, like melatonin, also influencing their amount of sleep.
A) This expression compares the share of college graduates who voted versus the share of college graduates who did not vote.
B) We would find the same estimate if we ran a regression of voting on if someone was a college graduate.
C) This expression compares the share of college graduates that voted versus the share of non-college graduates who voted.
D) This expression will give us the causal effect of being a college graduate on voting.
B & C
What does the the equation below represent?
A. This represents selection bias
B. This represents the naïve difference
C. This represents the causal effect
C (Causal Effect)