Research Methods & Data Analysis II Spring 2024 - LPPA 7160
Contact Information
Instructor | Email | Office |
Sebastian Tello Trillo | Garrett Hall 101 | |
Eileen Powell | ||
Ella Carlson |
Schedule & Office Hours
Lecture | Section 100 | TuTh 9:30am - 10:45am | Rouss Hall 403 |
Discussion | Section 101 | We 11:00am - 11:50am | New Cabell Hall 032 |
Discussion | Section 102 | We 6:00pm - 6:50pm | New Cabell Hall 368 |
Lecture | Section 200 | TuTh 11:00am - 12:15pm | Rouss Hall 403 |
Discussion | Section 201 | We 1:00pm - 1:50pm | New Cabell Hall 032 |
Discussion | Section 202 | We 5:00pm - 5:50pm | Minor Hall 130 |
Office Hours | Tello Trillo (in person) | Friday 10:00am-12:00pm
Sign up here | Garrett Hall 101 |
Tello Trillo (Virtual or in person) | |||
Eileen Powell | Monday 3pm-5pm | Pav 10 Basement Rm 2 | |
Ella Carlson | Tuesday 2:30-4:30pm | Pav 10 Basement Rm 2 |
Course Objectives
This is the second course in the quantitative methods sequence for graduate students in public policy. The sequence exposes students to the main empirical methods used in policy analysis. Our goal is to improve your ability to understand causal claims, critically consume empirical evidence, and communicate that evidence to policy audiences. The first half of this course will develop and extend your regression and data analysis skills. The second half will introduce experimental and quasi-experimental strategies for measuring policy impacts.
A byproduct of these objectives is to elevate the student’s skill in deductive reasoning and critical thinking. Specifically, we want students to feel confident about consuming complicated information and help them make a decisions based on that information.
Prerequisite: This course assumes a solid foundation in descriptive and inferential statistics, including probability, estimation, and hypothesis testing. LPPA 6150 is required.
This class will help you develop your quantitative toolkit. While we hope you find these tools valuable immediately, their full benefits may not become apparent for some time. Even if your future job doesn’t explicitly require statistics, the ability to analyze data and discuss evidence will enhance your performance no matter what path you choose.
We will try to minimize technical demands, relying on the minimum math and programming skills required to achieve course objectives. Still, some students will find the course’s analytical approach challenging.
Course Materials
Books
The following books are the main books for this class and available in the UVA Bookstore. Though they are not required, they will substantial help in your understanding of what we see in lecutre. The lectures will draw a lot from these books so they will be super helpful.
- Bailey, M.A. Real Stats: Using Econometrics for Political Science and Public Policy. New York: Oxford University Press. (B).
- There are two editions of this book. You are welcome to use either the first or second edition.
- Angrist, J. & Pischke, J-S. (2014). Mastering Metrics: The Path from Cause to Effect.
Princeton, NJ: Princeton University Press. (A&P)
You should first visit the books above for concepts and exercises. The Real-Stats book has questions and some answers in the back. However:
- If you have a concept that you need more or different perspectives on a particular concept
- If you need to see example coding
- If you need to see more “exercises” or “walk-through” examples.
Here are other books I recommend checking
- Gugerty, Mary Kay, and Dean Karlan. The Goldilocks Challenge: Right-fit Evidence for the Social Sector. Oxford University Press, 2018. (GK)
- This book is not made for “students” it is made for non-profits. They go over the concepts and importance of monitoring, when to monitor, and how to monitor and collect data for your non-profit. In addition, it suggest what to do when you cannot do an RCT.
- Stock, J.H. & Watson, M.W. Introduction to Econometrics. Boston: Addison-Wesley. (SW)
- An optional text for students interested in a more technical treatment of the course
- Cunningham, Scott. Causal Inference: The Mixtape V. 1.7. Obtain free at: https://mixtape.scunning.com. (C)
- This book is free and a great resource that combines intuition with higher-level math. This book is a bit more dense than the two required, but it is more fun too and has “different” examples. It also has some coding examples in STATA and R!
- Hernán MA, Robins JM (2019). Causal Inference. Boca Raton: Chapman & Hall/CRC. Obtain free at https://www.hsph.harvard.edu/miguel-hernan/causal-inference-book/ (H)
- Huntington-Klein, Nick. The Effect: An Introduction to Research Design and Causality (E) https://theeffectbook.net
material
Software
We will use Stata extensively in this course. Stata is available on the Hive virtual desktop. You can also purchase discounted versions using the Stata Corporation Campus GradPlan program. There are a number of good STATA resources online. UCLA, in particular, maintains a wealth of tutorials, FAQs, and videos.
Check out the resources page for more.
The course objective is not to teach you how to code, so this class is not designed to appropriately “teach coding”. We use coding as a way for the student to understand what’s happening with the data and how does one achieve the goals of causal inference. STATA is a user-friendly software that can quickly get everyone on board to understand what’s happening. Having said that, the student is welcome to use any other coding language to solve their homework, whether that’s R or Python, etc. The answer key and support from the instructors will be on STATA and that will be the course’s coding language.
Class Format
Much of the course will be in lecture format. We will frequently ask individuals to volunteer their thoughts, and we will also call on students to do so.
In general, you will not need cell phones during lecture. The available research suggests that they impede student learning both for computer users and their classmates.
When we use STATA examples during lecture, we will share the STATA files on the class website so that you may replicate the analysis on your own after class.
You should bring your computers to discussion sections, however. Those sessions will emphasize Stata and computer skills. If you believe your circumstances warrant an exception to these policies, please provide us with sufficient detail to evaluate your request.
Lastly, a note on notation: we will sometimes use different notation than the textbooks. This is intentional. Different authors use different notation, and we want you to be familiar with the alternatives.
How to approach lectures and course work?
To maximize learning, you should approach lectures as “review”. That means studying the concepts before coming to class through readings of Real Stats or Mastering Metrics, and then using lectures to review the concepts you’ve learned, and get to ask questions to understand better.
Each week students will have quizzes which will not factor in students grades to help with retention of the material. The goal of the quizzes are not to be evaluative, but to provide a way to practice active recall.
From time to time students will have worksheets. These are not required to be completed, but are a great way for the students to “touch” and play with the data to understand the methods that are being used.
Homework will be used to evaluate the student on how they are applying concepts and they will also serve as a didactic material to teach concepts.
Exams will mainly be used to evaluate progress and learning. However studying for exams has shown as a great tool to help student review the material.
Keys for Success
Over the years students have asked for keys for success to improve their learning and retention of the concepts learned in class. The following are a list of “tips” that different students have used:
- Read the assigned material before class. Lectures will cover difficult material and prior exposure will help. We will often assume that you understand basic background information from readings even if we do not cover that content in lecture. You will almost surely want to re-read the assigned material after class, especially if you found it difficult on a first read.
- Practice active reading. The best way to process the readings is with pen and paper. Econometrics is like reading math not history. But what should you write about? The usual is to make a summary of what you are reading. This is helpful because it helps with “retention” but copying and pasting concepts on paper will only be so helpful. The best type of summary notes are the ones in which your brain is processing the information. For example, instead of reading and taking notes as you read. Read the material and take very little notes. Then after you are done reading, try to summarize and organize the material in your own words without accessing the material, just from notes. A second tip when reading the material, is not to write a summary, but write a question that you need to know and then the answer. Then when you go back to notes, you can only see the question (Without the answer) and this can help with active recall.
- Practice active recall. Retention of the material is one of the hardest thing to achieve. This is because our brain are not designed for storage, they are designed to have ideas. You should aim to only retain core concepts and tools and then practice applying the basics on different problems. How to retain these core concepts? Active Recall is a tool to help with retention. The idea is to have a prompt (usually a question) for which you answer from what you remember and then you can see the answer. Here is a video explaining active recall and how to apply it.
- Apply things over and over. In order to understand your deficiencies you can apply concepts that you learn in class to your other classes, current events, and the policy questions that most interest you. This will help with (a) retention and (b) thinking about questions you have about the material. You can also use the materials we share from lecture and section to run analyses of your own. Come up with other applications for the tools that we cover. The Internet provides an unlimited supply of interesting data to explore.
- Weekly or biweekly review. Every week or two, make summary notes that highlight the key points from the lectures and relevant readings. Summary notes force you to make connections between material and assess what you don’t understand before exam crunch time hits.
- Allow yourself to struggle independently on assignments. Learn to seek out resources online to troubleshoot the Stata challenges that you encounter. Data skills and software are constantly evolving, so learning how to teach yourself new tools is more important than memorizing the ins and outs of any one program right now. A corollary: start assignments early. If you wait ’til the 11th hour, you will be tempted to cut corners and short change your own learning.
- Seek help sooner rather than later. There are many resources available to support your success in this course. Recall that when you seek help, try to figure out what exactly you need help with? Is it there readings? is it the applications? is it retention? What are you struggling with, and why?
- Check out the resources page for more resources!
Grading
Your grade will depend on homework and exams. The weighting scheme will be the following:
Homework | 60% |
Exam 1 | 20% |
Exam 2 | 20% |
Quizzes | P/F |
Homework
There will be approximately six or seven homework assignments. Some will emphasize the readings and statistical tools we cover in lecture. Others will require more substantial data analysis. For each assignment, we will specify whether you are permitted to work in groups or required to complete problems on your own. When group work is permitted, the group will turn in one assessment to be graded on.
The homework grade for the class-grade will be calculated by adding the points from each individual homework and then dividing them by the total possible points. This means that each individual homework won’t have a grade assigned (though each student is welcome to calculate that by dividing the points over total points of each homework).
Exams
There will be two exams. Missed exams will receive zero credit, except when a documented medical issue prevents you from taking the exam. Exams grade will be self-contained. That is each exam will have a grade over 100.
Regrading
A request for a re-grade shall be done within the next 5 days following receiving the assignment and answer key. Please check the answer key and submit in an email a description of why the regrade needs to occur. The re-grade will activate a re-grade of the entire assignment. It’s a simple process design for ease for the students.
Example
Monica grade’s are the following
Assignment | Grade | Total points |
Homework 1 | 23 | 25 |
Homework 2 | 57 | 59 |
Homework 3 | 60 | 60 |
Homework 4 | 40 | 45 |
Homework 5 | 90 | 93 |
Homework 6 | 12 | 15 |
Exam 1 | 45 | 50 |
Exam 2 | 63 | 67 |
Therefore Monica’s grade for the class will be calculated by the following
Monica’s Homework points: 282
Total possible Homework points: 297
Monica’s exam 1 grade would be
Monica’s exam 2 grade would be
Monica’s final grade would be
The mapping from numerical grade to letter grade will follow the standard assignment of As (with + and -) are in the 90s range, Bs in the 80s, etc. The final thresholds will be assigned at the end of the course.
Quizzes and the “progress policy”
Each week there will be a quiz available for students to practice the concepts learned that past week. These quizzes are not graded nor the factor into the final grade. However, if a student manages to get above 60% on their overall quiz grades (calculated by total number of correct questions over total number of questions), then that student will be able to use a “progress policy”.
The “progress policy” works the following way, the student will get 0.25 points of the positive difference between their Exam 2 - Exam 1. In other words, if a student has shown improvement over their exam, it can get points back on their first exam. This policy would only benefit students, it would not punish students (For example if Exam 2 < Exam 1).
Example. Let’s use the example above to see how this would change Monica’s grade. In Exam 1 the grade was 90 and in exam 2, 94. This means the improvement was of 4 points. Monica will get a quarter of those points back into exam 1, so the new final grade for exam 1 would be 91 instead of 90.
You will notice that this progress policy kick more points back in the larger the progress the student makes, this tends to then benefit people who didn’t do so well on midterm 1 and show great improvement in midterm 2.
Policy on Late Assignments
Late assignments will be accepted, but each late day will reduce your potential points by 10 percent — e.g., if you submit your homework three days late, you will receive 70 percent of the points you would have earned had you turned it in on time. Assignments submitted 10 or more days after the due date receive no credit.
Everyone is entitled to turn in an assignment 48 hours late without any penalty once a semester. Use it wisely. When you plan to use it, email your TA that you want your two-day-no-penalty to be used.
Frank Batten School of Leadership and Public Policy Honor Statement
The Frank Batten School of Leadership and Public Policy embraces and upholds the University of Virginia’s Honor Code principles that mandate that students will not lie, cheat, or steal, and we will not tolerate the actions of those who do. Acting in a manner consistent with the principles of Honor benefits every member of the Batten School community.
We expect every student to comply fully with all provisions of the UVA Honor System. By enrolling in this course, you agree to abide by and uphold the Honor Code System of the University of Virginia. The following applies to your Batten course work and requirements, and unless otherwise specified by your instructors:
- All graded assignments must be pledged.
- Students may not access any notes, study outlines, problem sets, old exams, answer keys, or collaborate with other students without explicit permission.
- When given permission to collaborate with others, students will not copy answers from another student.
- Students should always cite any resources or individuals they have consulted to complete an assignment. If in doubt, sources should be cited.
- Suspected violations will be forwarded to the Honor Committee, and, at the discretion of the instructor, students may receive “no credit" the assignment in question, independent of the actions taken by the Honor Committee.
- Any questions about what is or is not permitted on an assignment, should be clarified by students with their instructors prior to the completion of their work.
If you believe you may have committed an Honor Offense, you may wish to file a Conscientious Retraction (“CR”) by calling the Honor Offices at (434) 924-7602. According to Honor guidelines, for your retraction to be considered valid, it must, among other things, be filed with the Honor Committee before you are aware that the act in question has come under suspicion by anyone. More information can be found at www.virginia.edu/honor. If you have questions regarding the course honor policy, please contact your honor representatives.
Plagiarism
Because of misunderstandings among students about the definition of academic fraud, instructors are encouraged to address this subject in their classes and in their syllabi. Please distribute these standards in writing to the class and post in your course Collab site. You can find some good information about the honor system at UVA and resources for faculty and TAs here: https://honor.virginia.edu/faculty. The Batten School feels very strongly about academic fraud and enforcing the University honor code standards within our School. Faculty should take note that Batten supports the Honor Code and you.
Frank Batten School of Leadership and Public Policy Grading Policy
Due to increasing grade inflation in American higher education, the Frank Batten School has set a grade normalization policy, with a suggested grading distribution. All Batten courses should have a mean grade that does not exceed a 3.5 grade point average with an emphasis on a well distributed range of grades. If grades deviate significantly from this suggested distribution, a dean will discuss the course grades with the faculty member.
Frank Batten School of Leadership and Public Policy Well being Statement
If you are feeling overwhelmed, stressed, or isolated, there are many individuals here who are ready and wanting to help. Both Amanda Crombie, Director of Academic Programs and Jill Rockwell, Assistant Dean for Student Services are available to help all Batten Students. They are readily accessible during walk in hours or by setting up an appointment. Additionally, all Batten faculty and staff take student health and safety very seriously. Therefore, as part of their duty to care for distressed individuals, faculty and staff will refer students who threaten self-harm or suicide to appropriately qualified personnel at the earliest opportunity
Alternatively, there are also other University of Virginia resources available. The Student Health Center offers Counseling and Psychological Services (CAPS) for its students. Call 434-243-5150 (or 434-972-7004 for after hours and weekend crisis assistance) to get started and schedule an appointment. If you prefer to speak anonymously and confidentially over the phone, call Madison House’s HELP Line at any hour of any day: 434-295-8255. Additionally, as part of the student comprehensive health fees, you have 24/7 virtual mental health support through the TimelyCare app.
If you or someone you know is struggling with gender, sexual, or domestic violence, there are many community and University of Virginia resources available. The Office of the Dean of Students, Sexual Assault Resource Agency (SARA), Shelter for Help in Emergency (SHE), and UVA Women’s Center are ready and eager to help.
Class policies
Policy on infants in the classroom
You are welcome to bring your infant to the classroom as long as they are not distractive to others. Please consult me about a plan on infants.
Pronouns
If you have a preferred pronoun that I may not be aware of, please let me know.
Class Decorum
I aim to create an inclusive environment. If there is something about this environment that makes you feel not-welcomed. Let me know. I expect that you will attend class, having read all the assigned readings. A successful classroom community exists when we all come to class prepared and contribute to the class discussion in a thoughtful, critical, and active manner. Carefully listening to your classmates and building on their contributions will help facilitate a constructive, interactive classroom experience.
Please be respectful of your classmates’ thoughts and opinions. In this course we will explore a variety of socially sensitive topics. We will engage these topics directly but respectfully in order to explore important mechanisms and theories that underlie beliefs and behaviors with respect to policy domains. Consequently, this course will have an emotional impact on many of us. It will require an ability to listen and think, sometimes with as much dispassion as can be mustered. If you are willing to assume that everyone’s perspective is likely to have some merit, however, it is my belief that you will leave this course equipped with strategies for thinking about important and complex social issues.
In addition, please be professional in your communications. Be especially mindful of this over email, where tone is difficult to ascertain (e.g., sending me an email with “hey, I need the lecture slides” is not appropriate).
Important Dates
SPRING 2024 UVA Academic Calendar 2022-2024
January 16 Spring 2024 courses begin
March 2-10 Spring break – no classes
April 3-5 48 Hour Project for Introduction to Policy Analysis
April 30 Last day of classes
May 2-10 Exams (University Scheduled time)
DROP/ADD deadlines January 31 Final day to add a course February 1 Final day to drop a course, no grade penalty (course removed from transcript) March 13 Final day to drop a course with W on transcript (drop after this date will be F)
Course Schedule (Tentative)
Lecture Number | Week | Date | Topic | Readings | Assignments Due | |
1 | Week 1 | January 18 | Introduction & Motivation | |||
2 | Week 2 | January 23 | Lecture 2: Measurement & Size | Worksheets | ||
3 | Week 2 | January 25 | Lecture 3: Measurement & Size | |||
4 | Week 3 | January 30 | Lecture 4: Introduction to Causal inference with design | Potential HW Due Feb 5th | ||
5 | Week 3 | February 1 | Regression Mechanics 1 | Worksheets | ||
6 | Week 4 | February 6 | Regression Mechanics 2 | |||
7 | Week 4 | February 8 | Regression Mechanics 3 | Homework Due Feb 21st | ||
8 | Week 5 | February 13 | IV 1 | |||
9 | Week 5 | February 15 | IV 2 | Worksheets | ||
10 | Week 6 | February 20 | IV 3 | |||
11 | Week 6 | February 22 | IV 4 | |||
Week 7 | February 27 | Review Session | ||||
Week 7 | February 29 | Midterm (In class) | ||||
Week 8 | March 5 | Spring Break | ||||
Week 8 | March 7 | Spring Break | ||||
12 | Week 9 | March 12 | No Class | |||
13 | Week 9 | March 14 | Regression Discontinuity II: Concept // Fuzzy vs. Sharp | |||
14 | Week 10 | March 19 | Regression Discontinuity II: Concept // Fuzzy vs. Sharp | Homework Due | ||
15 | Week 10 | March 21 | Regression Discontinuity III: Assumption and aides | |||
16 | Week 11 | March 26 | Regression Discontinuity IV: Aides and Example | |||
17 | Week 11 | March 28 | Panel Data: Fixed Effects | Homework 4 (RD) Due Monday April 1st | ||
18 | Week 12 | April 2 | Panel Data: Fixed Effects | |||
19 | Week 12 | April 4 | 48 Hour Project (No class) | |||
20 | Week 13 | April 9 | DD 1 | |||
21 | Week 13 | April 11 | DD 2 | Homework 5 (FE) Due April 14th | ||
22 | Week 14 | April 16 | DD 3 | |||
23 | Week 14 | April 18 | DD 4 | |||
24 | Week 15 | April 23 | TBD: Putting all together part 1 Size of the effects, scaling, | |||
25 | Week 15 | April 25 | TBD: Putting all together part 2. Measurement | Homework 6 Due (DD) April 28th | ||
Week 16 | April 30 | Review Session | ||||
Friday, May 3 (Section 9:30am) | Final exam 2:00pm - 5:00pm | |||||
Thursday, May 2 (Section 11:00am) | Final Exam 9:00am - 12:00pm |