MPPA 610: Regression and Causal Inference in Policy Research

Spring I 2026 • Saturdays 9:00 a.m. – Noon • HyFlex (Armacost Library 151 & Zoom)

School of Business & Society • Master of Public Policy and Administration (MPPA)

Instructor

Professor Greg Thorson

Hall Chair in Public Policy and Political Science

Office: Hall of Letters 306

Office Hours: Saturdays, Noon – 2:00 p.m., and by appointment

Email: greg_thorson@redlands.edu

This course is offered in a HyFlex format. Students may participate in person in Armacost Library 151 or online via Zoom. The Zoom link and password are available on the course Canvas site.

Professor Greg Thorson

Course Description

This course is designed to expose students to the most commonly used advanced statistical techniques in the policy sciences, particularly those methods that are more likely to offer insight into whether relationships are truly causal rather than mere correlations.

We begin by reviewing the key concepts and procedures covered in your introductory statistics courses (up through linear regression), with a particular focus on the problem of endogeneity. After reviewing ordinary least squares (OLS) multiple regression, we move into advanced regression topics, including the use of dummy variables to test for interactions; regression with dichotomous dependent variables (e.g., logit and probit); nonlinear models (quadratics, polynomials, and logged variables); and interpretation and use of standardized coefficients.

We then cover a suite of cutting-edge causal inference tools commonly used in modern policy analysis. You will learn how and under what circumstances to conduct and interpret fixed effects models, difference-in-differences, regression discontinuity designs, propensity score methods, and synthetic control models.

Techniques will be explained both mathematically and conceptually. A strong command of math and calculus is not required for this class—“math phobes” are welcome. Like most professional social scientists, we will rely heavily on statistical software (Stata) to do the computations, while examining the underlying formulas and assumptions closely to understand both the strengths and limitations of each method.

By the end of the course, students will be fully prepared to engage with the next generation of causal inference techniques widely used in contemporary policy research, many of which are explored even more deeply in the second course in this sequence.

Course Learning Outcomes

Students who successfully complete the course should be able to:

Course Delivery Model

Students are responsible for their own learning and growth in the course. To support that learning, the course uses a combination of lectures, demonstrations in Stata, and discussion of texts and research articles.

It is the responsibility of each student to come to class fully prepared, having completed the assigned readings and exercises. Class sessions will combine brief lectures with extensive discussion. While the instructor will pose guiding questions, students will play a central role in leading discussion, critiquing methods, and interpreting results.

The class meets in HyFlex format: you may attend in person in Armacost Library 151 or online via the Zoom room linked from Canvas.

Required Texts, Hardware, and Software

Other Resources to Help You Learn Stata

Students who want additional support with Stata may find the following resources helpful:

Grade Basis and Grading

The grade basis for this class is numeric, ranging from 0.0 to 4.0. Final grades will be based on performance in the following areas:

Grade Scale

Course PercentGrade
93% – 100%4.0
88% – 93%3.7
83% – 88%3.3
79% – 83%3.0
75% – 79%2.7
71% – 75%2.3
67% – 71%2.0
63% – 67%1.7
60% – 63%1.3
57% – 60%1.0
54% – 57%0.7
Below 54%0.0

Learning Management System (Canvas)

The MPPA program uses Canvas as its official learning management system. Students are expected to check the course site frequently for updates and announcements.

Weekly Quizzes and Participation

Weekly participation and quizzes are central to your success in this course. Students are expected to:

Your participation grade is based on your presence, your contributions to discussion, and your preparation as demonstrated in both discussion and daily quizzes.

Class attendance and completion of weekly quizzes are required for successful completion of the course. If illness, emergencies, or other life events prevent you from attending, please contact the instructor as soon as possible to discuss possible accommodations (recordings, make-up quizzes, or alternative assignments, as appropriate).

Poll Everywhere Quizzes

We will use Poll Everywhere for weekly quizzes. Please download the free Poll Everywhere app for your phone or tablet, or access it via their website. Students generally experience fewer technical problems using the mobile app.

Log in using your .redlands.edu email address and join the presentation via the PollEv directory as instructed in class. Quizzes will cover:

Exercises

Exercises are designed to give you hands-on experience implementing the methods covered in class using Stata. You are welcome to work on exercises in groups; however, each student must submit their own work.

Exercises are due one week after we cover the associated chapter. For example, if we complete Bailey, Chapter 2 on January 13, then the exercise for that chapter is due at 11:59 p.m. on January 20.

Collaborative discussion of concepts and code is encouraged, but copying another student’s work or sharing files to be submitted by others is a violation of academic honesty and may result in failure in the course and further disciplinary action.

Current Research Studies

One of the goals of the course is to increase your statistical literacy, especially your ability to interpret and critique studies that appear in economics, public policy, political science, health, and related fields, as well as in the mainstream media.

At the beginning of each class, we will briefly critique recent quantitative research or statistical analyses from journal articles and media reports. On your assigned days, you are responsible for:

Examples may include:

The content does not need to be directly political. Mainstream newspapers, online news aggregators, and major think tanks (e.g., Brookings, AEI, Heritage) are all good sources. We will typically review 3–4 analyses each class period.

Students are assigned to groups for these presentations:

Office Hours

You are encouraged to talk with the instructor about course material, assignments, or broader questions about policy research. Office hours are:

To schedule an appointment, please use the online booking system: Book an appointment with Professor Thorson .

If no appointments are available, email the instructor or check back later, as additional slots are added regularly. Please arrive promptly to appointments and cancel at least 24 hours in advance whenever possible so that time can be used by other students.

University Policies and Student Support

Office of Equity and Title IX

The University of Redlands is committed to providing a safe and equitable learning environment. Discrimination, harassment, retaliation, sexual misconduct, and sexual harassment (including sexual assault, dating or domestic violence, and stalking) are not tolerated.

Many faculty and staff are designated as “Responsible Employees,” which means that if you disclose experiences related to these issues, they are required to report them to the Office of Equity and Title IX. You retain control over how you wish to proceed, including whether to pursue a formal complaint.

To learn more or to report directly:

Confidential resources include, but are not limited to:

For more information, visit the University’s Title IX and Equity website.

Writing and Subject Tutoring

Peer writing tutoring is available in-person and online and can be used at multiple stages of your writing process, in any discipline. You can schedule sessions at: https://redlands.mywconline.com.

Subject tutoring is available for many courses. Log in to the appropriate Subject Tutoring schedule through the same site, select your course, and choose a tutor and time. Tutoring is individualized and available up to two hours per week per course.

Counseling Center and CARE Team

The Counseling Center provides confidential mental health services, including individual and group therapy, single-session visits, consultations, and urgent appointments. Call 909-748-8108 or email counseling_center@redlands.edu. A 24/7 mental health crisis line is available at 909-748-8960. TimelyCare also offers virtual support.

The University CARE Team supports students who are overwhelmed, in distress, or may present a risk to themselves or others. If you have concerns about a fellow student, you can submit a CARE Team referral via the University’s online form.

Additional Student Support Resources

Names and Personal Gender Pronouns

It is important to the instructor to learn your preferred name and personal gender pronouns. If either is used incorrectly in class, please feel free to correct it during or after class. Your help in creating a respectful classroom environment is appreciated.

Freedom of Expression

The course is committed to the free expression of ideas. Particular viewpoints are not privileged simply because they are popular or supported by influential elites. All ideas and assumptions must be defended by reason and evidence.

If you enroll in this course, you should expect to encounter ideas you may find disagreeable, uncomfortable, or offensive. The classroom is not intended to be a “safe space” in the sense of avoiding discomfort. Rather, free expression should prompt you to re-examine your assumptions, values, and beliefs.

That said, there are limits: you may not threaten or harass others, nor may you attempt to suppress the views of others in the class. Students who make offensive statements should also expect those statements to be challenged by others who likewise enjoy freedom of expression.

For a fuller articulation of this approach, see the University of Chicago’s Report of the Committee on Freedom of Expression.

Accommodations

If you are a student with a disability requesting reasonable academic accommodations, please contact Academic Support and Accessibility (ASA), located in the Student Success Center on the ground floor of Armacost Library.

All requests for accommodations require registration with ASA in advance of need. Faculty, students, and ASA work together to determine appropriate classroom accommodations. You are encouraged to discuss approved accommodations with the instructor.

Academic Honesty

Scholastic dishonesty includes (but is not limited to) plagiarism; cheating on assignments or examinations; unauthorized collaboration; taking, acquiring, or using test materials without permission; submitting false or incomplete records; falsifying data or procedures; and misusing University academic records.

A student found responsible for scholastic dishonesty in this course may receive a penalty up to and including a grade of F for the course, and may also be subject to additional sanctions under University policy.

You are responsible for understanding and following all departmental and University policies on academic honesty.

Withdrawal and Incomplete Policy

Students are responsible for voluntarily withdrawing from the course if they decide not to complete it. If your name appears on the registrar’s final grade sheet and there is no work on which to base a grade, a grade of F must be entered.

University deadlines for dropping the course will be observed. Incomplete grades will be given only under extraordinary circumstances.

Course Schedule and Assignments

This schedule is tentative and may be adjusted at any time by the instructor. All changes will be announced in class and/or posted on Canvas.

Date Topic Readings and Media Assignments / Due Dates
January 10 Bivariate Ordinary Least Squares (OLS) Regression
  • Bailey, Real Stats, Chapters 1–3
  • Exercises: Chapter 3, Question 1
  • Due: January 17 at 11:59 p.m.
January 17 Multivariate OLS / Using Dummy Variables
  • Bailey, Chapters 4–6
  • Exercises: Chapter 4, Questions 1a and 1b; Chapter 5, Question 2a
  • Due: January 24 at 11:59 p.m.
  • Current Research Studies: Group 1
January 24 Nonlinear Models
  • Bailey, Chapter 7 and Chapter 12
  • Exercises: Chapter 7, Question 1; Chapter 12, Questions 1a, 1d, 1e (i and ii only)
  • Due: January 31 at 11:59 p.m.
  • Current Research Studies: Group 2
January 31 Fixed Effects and Difference-in-Differences Models
  • Exercises: Bailey Chapter 8, Exercise 1 (a, b, d, e only); Chapter 8, Exercise 4 (a, d, e only)
  • Due: February 7 at 11:59 p.m.
  • Current Research Studies: Group 3
February 7 Instrumental Variables / Regression Discontinuity
  • Exercises: Bailey, Chapter 11, Exercise 1 (a–g)
  • Due: February 14 at 11:59 p.m.
  • Current Research Studies: Group 4
February 14 Propensity Score Matching and Synthetic Control
  • Application and interpretation focused on matching and synthetic control.
  • Current Research Studies: Group 1
February 21 Interpreting Results from Causal Inference Research / Final Exam Distributed
  • Propensity Score Matching:
    Gleason, et al. (2014). “Do KIPP Schools Boost Student Achievement?” Education Finance and Policy, 9(1): 36–58.
  • Regression Discontinuity:
    Gray et al. (2023). “Employed in a SNAP? The Impact of Labor Requirements on Program Participation and Labor Supply.” Economic Policy, 15(1): 306–341.
  • Synthetic Control:
    Andersson, Julius J. (2019). “Carbon Taxes and CO2 Emissions: Sweden as a Case Study.” Economic Policy, 11(4): 1–30.
  • Take-Home Final Exam distributed in class and via Canvas.
  • Final Exam Due: March 3 at 11:59 p.m. (uploaded to Canvas).
  • Current Research Studies: Group 2

Reminder: All deadlines, updated dates, and any changes to the schedule will be announced in class and posted on Canvas. Students are responsible for monitoring Canvas regularly.