Applied Data Analysis

You can come to office hours to ask me questions about the course content (especially if you’re having trouble). But you can also come to office hours to say hello, ask me about my research, learn what political scientists do, and tell me about your interests (academic or otherwise). I’d love to meet you.

Course Description

Does ideological extremism cause a candidate’s vote share to increase? Does inflation cause presidential approval to decline? Does democratization cause countries to become more peaceful? These are all causal questions. They all take the form: does X cause Y?

As social scientists, we often care about causal questions, but it is difficult to answer these questions definitively. Unlike a pharmaceutical company that can randomly assign people to take or not take a drug, we cannot randomly assign ideologies to candidates, inflation to presidents, or democracy to countries (but it would be nice if we could!). Instead, we have to apply creative strategies to try to learn whether X causes Y with observational data. That is, data we observe from the real world (not an experiment).

In this course, we are going to learn about the social science toolkit that will help us analyze data from the real world and, sometimes, draw causal conclusions about whether X causes Y.

In a typical class, you can expect:

• Lecture on a new statistical/data analysis concept.
• In-class “lab” exercise where you will work in small groups to apply concepts we learned in the lecture.
• Review solutions to the lab exercise, answer questions.

By the end of this course, you should be able to:

• Distinguish correlation from causation, and understand how causal designs help us tell them apart.
• Select the right causal design for a given question.
• Interpret statistical output substantively and correctly.

Required Texts and Materials

All textbooks and materials used in this course are available for free online.

Our primary textbook is:

• Causal Inference: The Mixtape (referred to below as “Mixtape”) by Scott Cunningham (2021, Yale University Press).
  • Note: this textbook includes a good deal of mathematical notation. Because this is an applied course, our focus is on gaining a conceptual understanding of the math and implementing it with code and data. You are not expected to understand every equation on the first read.

I will also assign readings from:

• The Effect: An Introduction to Research Design and Causality (referred to below as “The Effect”) by Nick Huntington-Klein (2025, CRC Press).

This book covers many of the same concepts in a different, less technical way. You may want to reference this book if you get stuck or prefer its approach.

In-class labs, homework, and the final exam will require a basic understanding of the R programming language. We will cover everything you need to know about R on the first day of class. We will be using DataHub, which you can access with your UCSD Student SSO.

Course Format

This class is fully remote, and all sessions meet over Zoom (link on Canvas). Attendance, including turning your camera on, is required. Lectures will be recorded and posted to Canvas after class if you want to review anything.

If you experience a technical failure during class (your connection drops, Zoom crashes, your Wi-Fi goes out), try to reconnect as soon as you can. If technical problems are a recurring issue, please get in touch as soon as possible.

Assignments

• Attendance and Participation (15%) including attending lecture synchronously, turning your camera on, and actively participating in lab activities and discussions.
• Lab assignments (25%) completed in class and submitted on Canvas at the end of each session, graded for completion.
• Homework (30%) to be completed individually, applying key course concepts.
• Oral exam (25%) a short one-on-one conversation with me over Zoom in which you reason through a causal question.
• Final (5%) an asynchronous, comprehensive assessment completed individually.

Attendance and Participation (15%)

You are expected to attend every class session synchronously and have your camera on for the full session.

Because life happens, you will be allowed one excused absence with no questions asked. Missing one class will not impact your attendance grade, and the in-class lab from that day will be dropped from your final grade. You do not need to email me to take advantage of this policy.

Beyond your one free absence, you must contact me in advance if you need to miss additional classes. I may choose to grant you an additional excused absence, but that is at my discretion. If I grant an excuse, the lab from that day may be dropped or made up, depending on the situation.

Beyond your one free absence, missing class without prior contact (barring exceptional emergencies) will result in a zero on that day’s attendance and lab grades. Be thoughtful about how you use the free absence policy; I will not grant additional exceptions for routine reasons.

Lab Assignments (25%)

All class sessions include group lab activities, which allow you to practice what we have learned in a low-stakes, group setting. Lab groups will be assigned during the first week of class.

Labs are graded for completion, not correctness. Although these are group assignments, each member is responsible for submitting their own lab worksheet on Canvas at the end of the class session. Your separate participation grade (15%) covers attendance, keeping your camera on, and engaging in class, including the solution review below.

After lab, I’ll go over the solutions with the class. Solutions will also be posted after class so you can check your work. During this review, you may be called on to share or discuss your group’s approach. These are opportunities to learn from one another and share the thought process behind the work. These interactions will increase your participation score and cannot negatively impact your grade.

Homework (30%)

You will complete three homeworks over the course of the session, each covering material from the lectures and in-class labs. These include a series of multiple choice questions and will be completed and submitted on Canvas.

Each homework is graded for correctness, but it may be attempted up to three times. Only your highest score will count toward your final grade. My hope is that you work through the material and do not feel pressure to get every question right on the first try. I encourage you to use each attempt as an opportunity to revisit the lecture notes, textbook, and labs before trying again. Canvas will not reveal the correct answers until solutions are posted, one session after the deadline.

You must work on homework on your own. You may use our textbooks, notes, the internet, even AI. But you may not consult with other students, tutors, etc. You can also discuss your homework with me during office hours.

Oral Exam (25%)

During the last two days of class (July 27 and 29), each student will meet with me one-on-one over Zoom for a short conversation, roughly eight to ten minutes. You will be assigned a slot in advance. On July 27 and 29, you only need to join Zoom for your slot. There is no regular class session.

During your time slot, I will pose a short scenario related to the material we have covered, and we will work through how you would identify a causal effect in that situation. There is no coding or analysis; the exam is about reasoning, not execution. You will receive practice scenarios ahead of time, and I will say more about the format as we get closer to the end of the course.

Final (5%)

The course concludes with an asynchronous final exam, available from Friday, July 24 at 12:01 AM through Friday, July 31 at 11:59 PM. You may take it at any point during this window; budget two to three hours. The final is fully asynchronous: there is no in-person or synchronous exam, even though the window spans the university-assigned final exam slot. It resembles the homework, but is longer and comprehensive.

As with the homeworks, you must complete the final on your own, but it is open book: you may use your notes, the textbooks, the internet, or any other resources you find helpful.

Like the homework assignments, you will have three attempts on the final. The answer key will be provided after the July 31 deadline.

Course Policies

Academic Integrity

I take academic honesty and integrity seriously. You must adhere to the assignment-specific requirements in terms of what you may/may not consult in completing your work. Please see the UCSD policy on academic integrity for more information.

Use of AI

In this class, I encourage thoughtful use of generative AI tools (such as ChatGPT, Claude, Gemini, TritonGPT, etc). These tools are incredibly powerful and can help you with both the statistical and coding concepts taught in this course. However, over-reliance on these tools poses some risk. They can make you feel like you understand something more than you do, and while they are less likely to make explicit errors than in the past, they can mislead you if you do not understand the underlying concepts.

It is your responsibility to be a careful consumer of these tools and ensure that you validate anything you learn from them. It is also your responsibility, as a student in this course, to understand the answers they provide.

For some helpful hints on how to use AI in a thoughtful and responsible way, see my blog post Talking to AI.

Late Submissions

Labs cannot be submitted late. They must be completed and submitted during class.

Late homework will incur a one letter grade penalty for each 24 hour period it is late. For example, if a homework is due at 10:59am, a late submission delivered between 11:00am on the due date and 10:59am the following day will automatically lose one letter grade. All homework solutions will be posted one session after the homework is due. After solutions have been posted, late homework cannot be accepted.

Requests for Re-Grades

If you believe an error has been made, you have one week following the return of the assignment to request a regrade. After this point, re-grades cannot be requested. To request a regrade, please email Professor Noble with a brief explanation of why you are requesting a re-grade as well as evidence from our course materials justifying the request. I reserve the right to refuse to re-grade, and if I do re-grade, please note it may result in a lower grade.

Communication

For all questions or comments, you may get in touch with me during my office hours listed on this syllabus, or via email. If your email requires a response, you can expect one within 1–2 business days. If you email me over the weekend, the clock begins Monday morning.

Note: if you contact me the night before an assignment is due, I will not respond in time to provide any advice before the deadline. Please plan and work ahead.

Accommodations

Students needing accommodations for this course due to a disability must provide a current Authorization for Accommodation (AFA) letter issued by the Office for Students with Disabilities. Students are required to discuss accommodation arrangements with instructors and OSD liaisons in the department.

Other resources, including the inclusive classroom statement, advising, and resources to support equity, diversity, and inclusion, and more can be found in the Additional Resources section below the reading list.

Course Schedule and Readings

Based on your learning style, you may find it helpful to complete the readings before or after lecture. Ultimately, it is up to you when you want to do the readings. You can always refer to this syllabus for the most updated information about the course.

June 29 · Introduction and R Skills

• Slides
• Readings:
  • Does X cause Y? An in-depth evidence review by Holden Karnofsky.
• Optional:
  • For help with R: Data Science in R: A Gentle Introduction by James Scott, Chapter 1, Chapter 2, and Chapter 4.
• Homework 1 assigned.

July 1 · Potential Outcomes

• Slides
• Readings:
  • Mixtape, Chapter 4 (stop before “4.2 Randomization Inference”).

July 6 · Extraordinary Least Squares

• Slides
• Readings:
  • Jared Wilber, September 2022. Linear Regression. MLU Explain.

July 8 · Potential Outcomes (II)

• Readings:
  • The Effect, Chapter 10.

July 13 · Instrumental Variables

• Readings:
  • Mixtape, Chapter 7, sections 7.1–7.2, 7.3.1, 7.5.
• Optional:
  • The Effect, Chapter 19.
• Homework 1 due before class, at 10:59am PT.
• Homework 2 assigned.

July 15 · Instrumental Variables (II)

• No reading.

July 20 · Regression Discontinuity

• Readings:
  • Mixtape, Chapter 6, sections 6.1–6.2.3, 6.3.
• Optional:
  • The Effect, Chapter 20.
• Homework 2 due before class, at 10:59am PT.
• Homework 3 assigned.

July 22 · Difference-in-Differences

• Readings:
  • Mixtape, Chapter 9 (stop before “9.5 The Importance of Placebos in DD”).
• Optional:
  • The Effect, Chapter 18.

July 27 · Oral Exams, Day 1

• Oral exams begin; you will be assigned a slot in advance.
• Homework 3 due before class, at 10:59am PT. Please note this applies to all students irrespective of your oral exam time.

July 29 · Oral Exams, Day 2

• Oral exams continue.

Final Exam · Due Friday, July 31 at 11:59pm

Fully asynchronous; complete it on Canvas at any point in the July 24–31 window. See the Final section above for details.

Grading Scale

Additional Resources

These additional resources and the language come directly from the UCSD Political Science Department.

Inclusive Classroom Statement

I am fully committed to creating a learning environment that supports diversity of thought, perspectives, experiences, and identities. I urge each of you to contribute your unique perspectives to discussions of course questions, themes, and materials so that we can learn from them, and from each other. If you should ever feel excluded, or unable to fully participate in our class for any reason, please let me know, or please consult the Department’s Report an Issue page for additional campus resources to support you, and diversity, equity, and inclusion in our classroom, and beyond.

Additional resources to support equity, diversity, and inclusion in our classroom, and beyond, may be found here:

• https://diversity.ucsd.edu/
• https://students.ucsd.edu/student-life/diversity/index.html
• https://regents.universityofcalifornia.edu/governance/policies/4400.html

Resources to Support Student Learning

• Library Help, eReserves and research tools: https://library.ucsd.edu/ask-us/triton-ed.html
• Writing Hub: https://commons.ucsd.edu/students/writing/index.html
• Supplemental Instruction: https://aah.ucsd.edu/supplemental-instruction-study-group/index.html
• Tutoring: https://aah.ucsd.edu/content-tutoring/index.html
• Mental Health Services: https://caps.ucsd.edu
• Community Centers: learn about the different ways UC San Diego explores, supports, and celebrates the many cultures that make up our diverse community: https://students.ucsd.edu/student-life/diversity/index.html

Academic Advising

Students who have academic advising questions related to the Political Science major should contact the department’s Undergraduate Advisor, Zain Sharifi, via the Virtual Advising Center. Academic advising questions often include (but are not limited to): add/drop deadlines, course enrollment policies, planning major and minor requirements, quarter-by-quarter plans, department petitions and paperwork, and referrals to campus and student support services.

Equity, Diversity, and Inclusion Offices

Office of Equity, Diversity, and Inclusion. 858.822.3542 · diversity@ucsd.edu · https://diversity.ucsd.edu/

Office for the Prevention of Harassment and Discrimination. https://ophd.ucsd.edu/ · ophd@ucsd.edu · (858) 534-8298

UCSD Office of the Ombuds. https://ombuds.ucsd.edu/ · To reach a Confidential Ombudsperson, please call 858-534-0777.

UCSD’s Principles of Community

To foster the best possible working and learning environment, UC San Diego strives to maintain a climate of fairness, cooperation, and professionalism. These principles of community are vital to the success of the University and the well being of its constituents. UC San Diego faculty, staff, and students are expected to practice these basic principles as individuals and in groups. The Principles of Community and the Student Code of Conduct support equity, diversity, and inclusion in our classroom.

Acknowledgements

I thank Pamela Ban, Christopher Lucas, and Tiago Ventura for inspiration. I also thank Christopher Lucas for providing some of the materials that were used in this course.