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

Quiz 5

  1. Suppose you are estimating the effect of education on earnings, but you believe education is not exogenous. Instead, you use distance to the nearest college as an instrument for education. What concerns might you have about this instrument?
    1. Distance to the nearest college may affect earnings through local labor market effects
    2. Distance to the nearest college may make students more likely to attend college
    3. Distance and education are uncorrelated
    4. Distance and education might be strongly correlated
      5. ‣
      Answer

      A

  2. A researcher wants to estimate the effect of pretrial detention on conviction. In order to determine if the defendant is detained before the trial conviction the defendants get randomly assigned a judge for the judge to decide. This judge could be a lenient or a strict judge. This is a different judge that your ”conviction” judge. The research uses this judge assignment as an instrument for whether the defendant is detained before trial. Provide the option that correctly identifies, Y, D and Z
    1. D=Conviction, Z=pre-trial detention, Y=judge leniency
    2. Y=defendant’s guiltiness, Z=pretrial detention, D=judge leniency
    3. D=judge leniency, Y=conviction, Z=pretrial detention
    4. Y=conviction, D=pretrial detention, Z=judge leniency
      5. ‣
      Answer

      D

  3. Which of these equations identifies the reduced form?
    1. Conviction=β0+β1PreTrialDetention+ε\text{Conviction}=\beta_0+\beta_1\text{PreTrialDetention}+\varepsilonConviction=β0​+β1​PreTrialDetention+ε
    2. JudgeLeniency=β+β1PreTrialDentention+ε\text{JudgeLeniency}=\beta+\beta_1\text{PreTrialDentention}+\varepsilonJudgeLeniency=β+β1​PreTrialDentention+ε
    3. PreTrialDetention=β0+β1JudgeLenciency+ε\text{PreTrialDetention}=\beta_0+\beta_1\text{JudgeLenciency}+\varepsilonPreTrialDetention=β0​+β1​JudgeLenciency+ε
    4. Conviction=β0+β1JudgeLeniency+ε\text{Conviction}=\beta_0+\beta_1\text{JudgeLeniency}+\varepsilonConviction=β0​+β1​JudgeLeniency+ε
      5. ‣
      Answer

      D

  4. Which statement is the best justification for why judge assignment could be a valid instrument?
    1. Judge assignment guarantees that detention has a large effect on conviction
    2. Judge assignment is likely correlated with detention decisions, but as-good-as-random assignment may keep it unrelated to defendant characteristics in the error term
    3. Judge assignment is valid because it’s a judge not the defendant deciding detention
    4. Judge assignment is valid as long as the first-stage F-statistic is below 10
      5. ‣
      Answer

      B

  5. What is a definition of compliers in this setting?
    1. Defendants who get detained and because they are detained they are more likely to get convicted
    2. Defendants who are detained before trial regardless of whether they are assigned a strict or a lenient judge.
    3. Defendants who are detained if assigned to a strict judge, but released if assigned to a lenient judge
    4. Defendants who get released regardless of of whether they are assigned a strict or a lenient judge.
    5. Defendants who get convicted if they get detained
      6. ‣
      Answer

      B