Sebastian Tello
  • Home
  • CV
  • Contact
  • Research
  • Resources
  • RMDA
  • APP
InstagramBluesky

Quiz 10

  1. In 2018, several U.S. states raised their minimum wage while others did not. A researcher wants to estimate the effect of the minimum wage increase on teen employment (ages 16–19). She collects quarterly employment data from 2016–2020 for all 50 states, and uses DiD to compare states that raised their minimum wage in 2018 to those that did not. The researcher estimates the effect of the minimum wage increase using the following regression: TeenEmploymentqt=α0+β1WageRaiseqt+β2Post2018qt+δ(WageRaiseq∗Post2018t)TeenEmployment_{qt}=\alpha_0+\beta_1WageRaise_{qt}+\beta_2Post2018_{qt}+\delta(WageRaise_{q}*Post2018_{t})TeenEmploymentqt​=α0​+β1​WageRaiseqt​+β2​Post2018qt​+δ(WageRaiseq​∗Post2018t​), where:
  • “WageRaise” is a binary variable indicating 1 if its a state that had a wage raise in 2018 and 0 if not. (i.e., a panel observation from a state has a value of 1 in year 2016 if the state went on to raise its minimum wage in 2018).
  • “Post2018” is a binary variable indicating 1 if the panel observation is in a year after 2018.

      • What does δ\deltaδ capture?

    1. The difference in the change in teen employment from before to after 2018 for the treated states, minus the change in teen employment from before to after 2018 for the untreated states.
    2. The change in teen employment for treated states over time.
    3. The change in teen employment for untreated states over time.
    4. What would have happened to the control group had control states received treatment.
    5. The average teen employment between 2016-2017 for states that didn’t raise their minimum wage.
      6. ‣
      Answer

      A

  1. What does α0\alpha_0α0​ capture?
    1. The difference in the change in teen employment from before to after 2018 for the treated states, minus the change in teen employment from before to after 2018 for the untreated states.
    2. The change in teen employment for treated states over time.
    3. The change in teen employment for untreated states over time.
    4. What would have happened to the control group had control states received treatment.
    5. The average teen employment between 2016-2017 for states that didn’t raise their minimum wage.
      6. ‣
      Answer

      E

  2. What does α0+β2+δ\alpha_0+\beta_2+\deltaα0​+β2​+δ capture?
    1. The difference in the change in teen employment from before to after 2018 for the treated states, minus the change in teen employment from before to after 2018 for the untreated states.
    2. The change in teen employment for treated states over time.
    3. The change in teen employment for untreated states over time.
    4. What would have happened to the control group had control states received treatment.
    5. The average teen employment between 2016-2017 for states that didn’t raise their minimum wage.
      6. ‣
      Answer

      D

  3. The researcher plots employment trends for treated and control states from 2016–2017 (pre-treatment). She finds they on average the control states had higher employment than the treated states, but they were trending closely together. Why is this relevant for DD?
    1. It confirms that treated states had higher employment before the policy.
    2. It provides evidence in support of the parallel trends assumption.
    3. It shows that we cannot use DD because treatment and control are not similar in a particular relevant observable characteristic.
    4. It shows the two groups are identical in all characteristics.
      5. ‣
      Answer

      B

  4. Suppose that in 2017, an economic recession hit only the states that would later raise their minimum wage. This will not create bias in our main DD assumption:
    1. True, It would have no effect since the recession happened before the treatment.
    2. False, It could violate the main DD assumption because it would create different trends before the policy
      3. ‣
      Answer

      False