final-practice.spr08.ps3 - PS3: Practice Questions for...

Info iconThis preview shows pages 1–3. Sign up to view the full content.

View Full Document Right Arrow Icon
PS3: Practice Questions for Final Exam Note. These questions illustrate the style of questions you will see on the final exam. You may see topics on the final that are not reflected in these practice questions, and vice versa. Do not take these questions as an exhaustive reflection of topics you should select to study. Your mileage may vary, use at your own risk, questions come with no warranty express or implied, etc. 1
Background image of page 1

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
1. Oregon has an experimental prison boot camp program to rehabilitate prisoners before their release. The object is to reduce the recidivism rate — the percentage who will be back in prison within three years. Prisoners volunteer for the program which lasts several months, but many prisoners drop out before completing the treatment. To evaluate the program, policy analysts compared prisoners who completed the program with prisoners who dropped out. The recidivism rate for those who completed was 29%, and for the dropouts it was 74%. Based on these findings, the policy analysts concluded that the program worked. (a) Was this an observational study or a controlled experiment? Explain your answer. (b) Do the dropouts provide good information about the counterfactual outcome of the completers? (c) Why might another analyst be skeptical that the program caused the dif- ference in recidivism? (d) Another policy analyst evaluated the program with a different methodology. She combined program completers and dropouts into one group, which had an overall recidivism rate of 36%. She compared this combined group with the prisoners who did not enroll in the boot camp program, who had a recidivism rate of 37%. Do these findings support the claim that the program works? Or that it has no effect? Explain. 2. (a) Define average treatment effect. (b) Define quasi-experiment. (c) Define conditional independence (of treatment and covariates). (d) True or false, and explain: if conditional independence does not hold in a quasi-experiment, it is not possible to compute the difference in average outcomes for treated and untreated units. 3. A snippet of data from the National Health Interview Survey (performed by the U.S. Census Bureau and sponsored by the National Center for Health Statistics) is shown in the table. Thus, for example, 70% of the people age 18-64 ate breakfast every day over a certain, compared to 90% of the people age 65 and over. Age
Background image of page 2
Image of page 3
This is the end of the preview. Sign up to access the rest of the document.

This note was uploaded on 05/20/2008 for the course MATH 16A taught by Professor Stankova during the Spring '07 term at University of California, Berkeley.

Page1 / 5

final-practice.spr08.ps3 - PS3: Practice Questions for...

This preview shows document pages 1 - 3. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online