Econometrics for Policy Research II

Course Logistics

  • Wednesdays, 6:10-8:00, 1957 E, Room 309
  • Lab: Wednesdays, 8:10 - 10, GELM B01

Course Description and Objectives

The essence of research is posing a question about how the world works, generating a hypothesis, and using evidence to test that hypothesis. This course focuses on how we use evidence to test hypotheses.

For masters' students, you should leave this course as an informed consumer of research. Is the research design you're reading credible? Why or why not? What are red flags in published research that should make you look more carefully? What are the steps required to generate a credible research product? What is causal inference, and how do you evaluate causal strategies?

For PhD students, my additional goal is to teach you how to understand estimation techniques. You may or may not use the techniques we learn in this course. Regardless, understanding their logic will help you interpret and apply empirical techniques in your own work.


This course is the second in a two-part graduate sequence in econometrics. It follows the content from PPPA 6013.

As a result of completing this course you should be able to
  • Apply reduced-form empirical techniques to research questions you develop
  • Replicate the data assembly and analysis in a published paper
  • Know how to apply the tools of policy evaluation
  • Differences-in-differences
  • Instrumental variables
  • Regression Discontinuity
  • Matching
  • Read and critique causal arguments in academic papers
  • Use statistical software to implement the tools of policy evaluation

Contact and Office Hours

Professor: Leah Brooks
Media and Public Affairs Building, Room 601F
Office Hours: Tuesdays 10 AM to 1 PM.
Tuesdays, 1:15 to 4:45 pm (last appt ends at 5 pm)
Thursdays, 8:15 to 9:15 (last appt ends at 9:30 pm)
Use the scheduler to book these times.
Due to the virus, I also have consultations during class times not otherwise allocated. Sign up here lfbrooks at gwu.edu
202-994-4703

Contact policy: I will do my best to answer emails within 24 hours during weekdays, or within 24 hours on the soonest weekday if you email on the weekend. If you do not hear from me within this time frame, you should assume that your email has been lost and you should re-send.

If you have missed a class, your first line of defense to ask what you have missed is another student.

If you cannot make it to office hours in person, I am happy to talk on the phone or via google hangout. If you want to reach me by phone, please just call at the time you have scheduled. If you'd prefer to use google hangout, please let me know in advance and I will be online.


Graduate Assistant: Jim Kim
jkim12 at email.gwu.edu
202-830-5829
Office Hours: Wednesdays 8 to 10 during lab hours in Gelman B01. Otherwise by appointment.

Jin will answer emails within 24 hours, except for those periods when he may be traveling.  He will notify the class about these periods in advance.

Prerequisites

PPPA 6013: Econometrics for Policy Research I

This class is substantially more difficult than PPPA 6013 and requires either familiarity with statistical programming, or the ability and willingness to learn this skill while taking the course. Please see me to discuss if you are unsure whether this course is appropriate for you.

This class also uses, but does not directly teach, Stata (you are also welcome to use any statistical package of your choice if you prefer). If you would like to learn some Stata on your own, I highly recommend the UCLA website, which has many self-directed tutorials. If you complete the “Fundamentals of Stata 1 and 2” and the two data management tutorials herehere and this this old (but good -- all the basic commands remain the same) tutorial, that is more than sufficient background to proceed in this class.

Do not do these tutorials with the expectation of mastering Stata. What you need to do is to learn enough to know diagnose problems and find solutions. This will give you the tools to find solutions for problems outside of material that you've learned directly.

Readings

Required textbook: Angrist, Joshua D. and Pischke, Jorn-Steffen. Mastering Metrics.

The textbook is not on order at the campus bookstore, but is widely available online. I have tried to link to all remaining readings from this syllabus. Please let me know if you have difficulties with any of the links, or with permissions.

Readings are subject to change, given the pace at which we move through the material.

Software

You are not required to use Stata for analysis in this course, but it is the program for which the TA and instructor can provide support.

Your options to access Stata are to use it in the GW library, use it on GW's cloud, buy a student version, or get an account on GW's high performance computing cluster (only feasible if you have some familiarity with Unix/linus and vi/emacs/other text editors). The TA will discuss these options during lab.

Course Overview

C#     Date Topic Due wk/oh
1 January 15 Causality and Regression Review n/a
2 January 22 Fixed Effects wk
3 January 29 Diff-in-diff 1 proposal and PS 1 wk
4 February 5 Diff-in-diff 2 oh
5 February 12 Instrumental Variables 1 Problem Set 2 oh
6 February 19 Instrumental variables 2 oh
7 February 26 Regression Discontinuity 1 oh
8 March 4 Regression Discontinuity 2 Problem Set 3 wk
9 March 11 Matching 1 Quantitative Progress n/a
March 18 Spring Break n/a
10 March 25 Matching 2 and Paper Workshop oh
11 April 1 Consultations Presentations oh
12 April 8 In-class workshop Presentations oh
13 April 15 Consultations Presentations oh
14 April 22 Structural Estimation
April 24 Paper Due

During workshops (wk), the TA teaches and reviews Stata skills. In weeks without workshops, the TA is available for office hours (oh).

The plan for Stata workshops is as below. If you have additional requests, speak with Jin. Note that this schedule is subject to change in Jin needs to fly out for job interviews.


C#     Date Topic
2 January 22 Basic Stata overview, including merging and collapsing
3 January 29 xi regression and fixed effect variables
8 March 4 How to export tables/regressions/charts to Excel
10 March 25 Paper Workshop

Evaluation

  1. Problem Sets (20%)
    • The three problem sets are designed to practice the skills we learn in class
    • And to help you prepare with Stata for writing the paper
    • Turn them in at the beginning of class that they are due
    • Any problem set turned in after that receives half-credit
    • Problem sets should be typed
    • You're welcome to work with others, but you should each turn in your own work, in your own words
  2. Research Paper (60%)
    • 10 to 15 pages; no more than 15 pages
    • Paper is due at the final class, in class
    • Extensions will be given only the case of illness
    • Essays will be graded out of 100 points
    • Any essays submitted late will decline by ten points for each twelve hours the essay is late, e.g. if the essay is due on Friday and is received Monday, if it would have received 70%, it now receives 30%
    • To make sure you are on track, we have two way-markers that each count for two and a half percent of the paper grade
      • A proposal due January 29 (lecture 3)
      • Evidence that you've made progress on the quantitative front, due March 25 (lecture 10)
      • In-class workshop, where you comment on drafts, April 1 (lecture 11)
      • Late work for these way-marker projects receives a grade of zero
  3. Paper Summaries (10%)
    • For the semester, each of you will write three paper summaries
    • I'll randomly assign you to a week; feel free to trade weeks amongst yourselves and let me know.
    • Write a one page summary of the paper we are discussing that week. At least a third of the summary should be a critique or extention of the article.
  4. Class Participation (5%)
    • I expect that you will come to class having done the reading and that you will be prepared to engage with me and other students in discussing the material
  5. Research Paper presentations (5%)
    • Comments on your classmates' presentations (2.5%)
    • Your presentation (2.5%)

Trachtenberg School Course Policies

  • The Syllabus
    This syllabus is your guide to the course. If any questions arise, please check the syllabus before contacting me or the TA. Sound educational practice requires flexibility and I may revise content and requirements during the semester.
  • Late or Missed Class
    If you are late or absent from class, it is your responsibility to obtain all announcements, assignments, and handouts from this website or from your classmates. As participation is part of your grade, and because attendance in class helps you learn, missing many classes will be detrimental to your final grade. Missing one class should have no effect. You do not need to notify me in advance if you are going to miss class.
  • Due Dates
    Please notify me in the first two weeks of class if you are aware of a pre-existing conflict, such as a religious holiday you observe, that will preclude you from completing work at the assigned time. To the extent possible, we will work together to reschedule as close to the original date as possible. As most work in this class is given with substantial advance notice, I am hopeful that re-scheduling is unlikely to be necessary.
  • Submission of Written Work Products Outside of the Classroom
    It is your responsibility to ensure that I receive your assignment on time. It is not permissible to submit assignments digitally unless I indicate so.
  • Collaboration on Assignments
    You are welcome to work in groups; however, you are expected to write up your answers individually. This means that no phrases on your assignment should mimic phrases on any other student's work.
  • Submission of Written Work Products after Due Date
    All work must be submitted by the assigned due date in order to receive full credit. Only extreme circumstances warrant exceptions. Late assignments will be marked down for each day that they are late.
  • Academic Honesty
    All examinations and other graded work products are to be completed in conformance with the George Washington University Code of Academic Integrity. Note especially the definition of plagiarism: “intentionally representing the words, ideas, or sequence of ideas of another as one's own in any academic exercise; failure to attribute any of the following: quotations, paraphrases, or borrowed information.”
  • Incompletes
    You must consult with me to obtain an incomplete no later than the last day of classes in the semester. At that time, we will both sign the CCAS contract for incompletes and submit a copy to the School Director. Please consult the TSPPPA Student Handbook or visit this link for the complete CCAS policy on incompletes.
  • Changing Grades After Completion of Course
    No changes can be made in grades after the conclusion of the semester, other than in cases of clerical error.
  • Accommodation for Students with Disabilities
    If you need extra time on exams or assignments due to a disability, let me know in the first week of class. In order to receive accommodations on the basis of disability, you'll need to provide proper documentation to the Office of Disability Support Services, Marvin Center 436, 202-994-8250. Accommodations will be made based upon the recommendations of the DSS Office.
  • University Counseling Center
    The University Counseling Center (UCC), 202-994-5300, offers 24/7 assistance and referral to address students' personal, social, career, and study skills problems. Services for students include: crisis and emergency mental health consultations; confidential assessment, counseling services (individual and small group), and referrals
  • Religious Holidays
    If you need to miss a class to observe a religious holiday, please notify me the first week of classes about any conflict; we will arrange an absence without penality.
  • Out of Class Learning
    Average minimum amount of independent, out-of- class, learning expected per week: In a 15 week semester, including exam week, students are expected to spend a minimum of 100 minutes of out-of- class work for every 50 minutes of direct instruction, for a minimum total of 2.5 hours a week.