Homework answers / question archive / 1

1

Computer Science

Share With

---

1.) Imagine the following: In the 1960s all men of military age (18+ years) became draft eligible to fight in the Vietnam War. A lottery is used where the social security numbers of draft eligible men were randomly drawn. Those chosen would have to serve in the military.

Q1: for which of the following could the random lottery draft be used as a natural experiment?

 

A.) To study the effects of military service on long-term mental health
B.) To study why people choose to join the military
C.) To study whether drafts are effective tools of recruitment or ultimately counterproductive

 

2.) Now imagine that, contrary to official policy, people could pay unscrupulous doctors to produce reports of illnesses and disabilities that would *disqualify* someone from the draft.

 

Q2: In which of the following ways would this fact mess up the natural experiment? Choose one

 

A.) The fact that many people can evade the draft will reduce the number of observations and increase our uncertainty about the effects we estimate
B.) Those who are drafted and not drafted are no longer as-if random, since the unscrupulous doctor is a backdoor path
C.) The fact would not undermine the natural experiment, since a backdoor path must affect both treatment and outcome

 

3&4) Read the abstract (and if you want, the introduction) to this paper by Rafael Di Tella and Ernesto Schargrodsky (2003), which uses a difference-in-difference design: https://bibliotecadigital.fgv.br/dspace/bitstream/handle/10438/13019/000317610_s311d.pdf?s

Q3: What is the treatment group, control group, and the pre- and post-treatment periods in this paper?

Q4: What would it mean, in this paper, for the parallel trends assumption to hold? Choose one

 

A.) The rate of Jewish institutions in the places affected and unaffected by terrorism would look similar prior to treatment onset
B.) The rate of robberies in places with and without Jewish institutions would look similar prior to treatment onset
C.) The rate of terrorist attacks in places with and without robberies would look similar prior to treatment onset