University of Michigan's Predoctoral Training Program in Causal Inference in Education Policy Research

We train doctoral students to become experts in causal inference in education policy. Our goal is for every fellow to have a deep expertise in her specialized set of methods, as well as sufficient understanding of the foundational methods of causal inference so that she can work productively in partnership with others who complement her expertise.

Susan Dynarski, program director

Program Overview

The University of Michigan School of Education, Gerald R. Ford School of Public Policy and Department of Economics offer a doctoral training program to students interested in using causal methods to evaluate educational policies and practices spanning early childhood to the labor market. The program builds on the University of Michigan's strengths in education policy and assessment in the social sciences and contributes to the University's emphasis on interdisciplinary research and educational programs that make critical contributions to society.

Fellows participating in the program will study a set of required courses that provide formal training in quantitative methods and contextual knowledge about education policy and practice. They will participate in research apprenticeships that apply these concepts to a research project under the supervision of core faculty. Fellows will regularly participate in the Causal Inference in Education Research seminar, a research workshop at which doctoral students and faculty present their own research and constructively critique the research of their peers. In the research colloquium, fellows will be introduced to eminent scholars who use the methods taught in the training program to develop the fellows' substantive knowledge and professional network. Finally, fellows will receive training and coaching in professional skills, including developing and sustaining partnerships with practitioners; writing and presentation skills for both academic and practitioner audiences; grant-writing and grants administration.

The program is led by professor Susan Dynarski, who is joined by professors Brian Jacob, Stephen DesJardins, John Bound, Jeff Smith and assistant professor Christina Weiland. The program is supported financially by a $4.0 million grant from the Institute of Education Sciences (grant R305B150012) as well as $1.5 million by the University of Michigan's Rackham Graduate School, Gerald R. Ford School of Public Policy, School of Education, Department of Economics, and the College of Literature, Science and the Arts.

Courses

Required coursework will provide formal training in quantitative methods and contextual knowledge about education policy, institutions and practice. Many courses will satisfy both degree and training program requirements. The sequence of required courses includes:

Five courses in Quantitative Methods

The Quantitative Methods requirement provides fellows with the formal technical training they need to critically consume and thoughtfully produce quantitative research about education policy. The program requires a total of five quantitative courses, as follows:

  • Two (2) foundational courses in statistics, covering topics up to multiple regression analysis. Prospective fellows will complete these courses before applying to the training program.
  • Three (3) additional courses in advanced quantitative methods.

Two courses in Causal Inference in Education Policy Research

​The year-long sequence of courses in Causal Inference in Education Policy introduces students to education research that employs causal methods, and provides institutional, historical and theoretical context for the questions addressed in this research. Courses will focus, not just on identifying average causal effects, but also on determining the mechanisms by which effects were obtained, measuring intervention fidelity, and detecting heterogeneity in treatment effects. Fellows will take the year-long sequence in the first year of their fellowship.

Education 712 / Public Policy 712 Causal Inference in Education Policy Research: ​Preschool, ​Elementary and Secondary​
Education 714 / Public Policy 713 Causal Inference in Education Policy Research: Postsecondary​

The sequence ​exposes students to the fundamentals of applying causal methods to education research, covering several techniques:

  • randomized controlled trials
  • regression discontinuity design
  • instrumental variables
  • fixed effects
  • differences-in-differences
  • matching

One course in Education Policy, Institutions or Practice

Fellows will take one additional course that strengthens their knowledge of education policy, institutions or practice. There will not be a list of approved courses. The only proscription is that the course should not focus on quantitative methods, since its intent is to deepen fellows' knowledge of the context in which education policy is made and operates.

Apprenticeships

Fellows will participate in a research apprenticeship with core faculty in each year of their fellowship, with at least one year on a project that is conducted in partnership with an education practitioner or policymaker. While working on research apprenticeships, fellows will participate in every phase of the research process. They will work with faculty on developing research questions, reviewing literature, choosing an appropriate methodology, and applying it to data. They will work with student-level, longitudinal, administrative datasets and will develop expertise in coding and statistical analysis. They will present initial results, work on presentations and paper drafts, and attend conferences where results are disseminated.

Examples of research questions addressed by projects currently led by the core faculty:

  • What is the effect of a universal pre-kindergarten program on student learning and social development?
  • Does offering dual-credit math courses to high school students improve their rates of postsecondary attainment and success?
  • What is the effect of online classes, compared to face-to-face learning, on the academic achievement of secondary students?
  • Does enrollment in developmental math increase or decrease the likelihood that community college students will study and earn a credential in a STEM field?

Select Research Projects

Early Education

K-12 Education

Postsecondary Education and the Labor Market

Other Topics in Education Research

Faculty and Staff

The program is led by Susan Dynarski, professor of education, public policy and economics, and is housed at the Education Policy Initiative within the Ford School. Dynarski is joined by faculty at the Ford School, School of Education and Department of Economics, with expertise in domains ranging from early childhood to the labor market. Core faculty will teach core courses, participate in the fellow selection process, mentor and advise fellows, manage the fellows' research apprenticeships, and work closely with fellows to ensure their success.

Core Faculty

Susan Dynarski, Professor of Education and Public Policy and Economics
John Bound, George E. Johnson Collegiate Professor of Economics
Steve DesJardins, Professor of Education and Public Policy
Brian Jacob, Walter H. Annenberg Professor of Education Policy, Professor of Economics and Professor of Education
Jeff Smith, Professor of Economics and Professor of Public Policy
Christina Weiland, Assistant Professor of Education

Staff

Mahima Mahadevan, Research Manager
Julie Monteiro de Castro, Program Manager and Administrator

Benefits

The training program will include 3-year and 4-year fellowships. Fellows supported financially by the program will receive an annual stipend of $30,000, full tuition support, fringe benefits and a small research allowance to attend academic conferences in each of year of their fellowship. Funding is currently available through June, 2020.

Students who successfully complete the program will be in position to pursue many careers, including within academia, research organizations and leadership positions within state and federal education agencies.

Admissions

The program will admit its first cohort in Fall 2016. Applications will be solicited from doctoral students across the social sciences at UM. The program seeks to build a diverse pool of potential fellows, including highly qualified students from educationally-disadvantaged backgrounds, underrepresented cultural and ethnic groups as well as students with disabilities.

Eligibility

Students must be enrolled full-time in a UM doctoral program in the social sciences. Fellows must be citizens or permanent residents of the United States to receive funding through the training program. Fellows must conduct independent research as well as dissertation research related to education. Renewal of funding in subsequent years is conditional on meeting annual training program requirements.

Selection Criteria

Admissions decisions will be based on the candidates' demonstrated interest in the content of the training program, academic performance, and faculty recommendations. Interested applicants are strongly encouraged to attend CIERS to demonstrate and test their interest in the training program. Candidates must show strong performance in quantitative analysis (e.g., grades in previous coursework, GRE scores) in order to gain admission to the training program. A fellowship committee will review applications and select the fellows.

Offers of fellowships will typically be made in the spring of candidates' first or second year of doctoral studies. Students will enter the training program in their second or third year of doctoral studies.

How to Apply

The program is not currently accepting applications. Check back in academic year 2016-2017 for further details.

Process

The application process will be conducted in two stages, as follows.

In the first stage, all applicants will provide, via online application​, a statement of interest, CV,​ ​​and access to their original PhD application and current​ UM transcript.

Applicants selected for a second stage will receive an invitation to submit one reference ( current UM faculty) and to meet with members of the fellowship committee. 

Statement of Interest for the Causal Inference in Education Policy Research Fellowship

  1. Please describe your academic and professional experience as it relates to education policy research and/or causal inference methods. (maximum of 250 words)
  2. Please describe your area(s) of interest in education policy research and the core faculty member(s) with whom you would most like to work. (maximum of 250 words)

Timing

The online application will be open between January 11 and March 11. Applicants will be notified of offers by April 22.

Contact

For more information, please email um-edpolicy-predoc@umich.edu.

We strongly encourage all students who are interested in the program to attend the weekly Causal Inference in Education Research seminar to better understand the types of research and discussions in which our students and faculty engage.

General Inquiries

For all general inquiries about the Education Policy Initiative, please use the contact form.

Street Address

Education Policy Initiative
Gerald R. Ford School of Public Policy
735 South State Street, Suite 5100
Ann Arbor, MI 48109

Get in Touch

Phone: 734-615-6978