SIEPR UGRA Summer Program: Open Positions
Summer UGRA Program
SIEPR offers a 10-week full-time (40 hours per week) Summer research program for undergrads, from June 26- September 1, 2023. Students work on a faculty-led research project and also participate in weekly seminars and meetings with their peers. Each student receives a stipend of $7,500 for the period. SIEPR aims to introduce a diverse population of students to economic policy research and encourages students from all areas of study to apply.
SIEPR UGRA Student Eligibility:
- Participants must be current Stanford undergraduates.
- Coterm students and seniors are eligible only if their bachelor’s degree will not be conferred before the end of the research appointment.
- Coterm students paying graduate tuition are not eligible.
- Students serving a suspension or on a leave of absence during Summer quarter are not eligible.
Program Activities:
- Students participate on campus.* There is no remote work option.
- Students work full-time on a faculty-led project throughout the summer.
- Students engage in an hour-long in-person working group meeting each week on Wednesdays to discuss their research progress with their peers.
- Students attend an hour-long, in-person seminar or professional development activity each week on Fridays to learn about different faculty or graduate student research projects, and other research opportunities.
- Students will work in teams to develop a policy project. In Week 10, each student presents a lightning talk about their part in the policy project and submits their talking points on slides at the final working group meeting.
Application Requirements:
- A list of the courses that you have taken that are research-related.
- Resume
- A cover letter that addresses the following:
- Why are you interested in a SIEPR UGRA position?
- What is your previous experience, if any, with research?
- What are your personal research interests?
Current Projects
Accelerated Learning in Community Colleges
Faculty Mentor: Eric Bettinger
Recently many universities and colleges are employing a term structure on top of existing semesters. Schools offer accelerated courses in 7-8 weeks rather than the full 16 weeks. We estimate the impact of this program on students' academic outcomes.
RA Responsibilities: Clean data. Run regressions. Help code interviews.
RA Qualifications: Basic econometrics. Ability to organize and create structure in analyzing interview transcripts
Language Models for Judicial Behavior Detection
Faculty Mentor: Daniel Ho
What implications does the rise of ChatGPT and foundation models have for law? The purpose of this project is to (1) construct legal foundation model benchmarks from existing studies of judicial behavior, and (2) evaluate the performance of foundation models (FMs) on these benchmarks. Studies of judicial behavior studies focus on describing and characterizing the decision-making processes and outcomes of courts, often with respect to the legal tools applied, issues litigated, or properties of the parties. To do so, legal scholars manually annotate collections of judicial opinions and other legal documents. As part of this project, we intend to accumulate a representative collection of judicial behavior studies (along with their annotations), develop a typology to define different types of studies, and evaluate whether modern FMs can automatically perform these annotation processes. The ultimate goals of the project are to assess the performance of FMs in the highly specialized domain of law, and determine if FMs can constitute a novel tool for empirical legal research. If FMs can indeed be used to automotive large scale analysis of legal texts, they signal a significant reduction in the costs of accessing legal knowledge.
RA Responsibilities: Build dataset of judicial behavior studies (along with their annotations), develop a typology to define different types of studies.
RA Qualifications: Ideal candidates for this position should have an interest in law/computational social science and familiarity with language modeling.
Political Networks, Polarization and Democratic Backsliding
Faculty Mentor: Saumitra Jha
Analyzing novel data on networks of influential individuals, from Oathkeepers in the US to Revolutionary War veterans to the French Resistance
RA Responsibilities: Collected and analyze novel data on networks of influential individuals, from Oathkeepers in the US to the French Resistance
RA Qualifications: Interest in and willingness to read original historical sources; knowledge of French is also a plus
State and Local Economic Policy Research
Faculty Mentor: Joshua Rauh
The RA in this role will aide a team of researchers who focus on state and local economic policy, including pensions/debt, economic development, labor force development, energy, and other topics.
RA Responsibilities: Data cleaning, data analysis, quantitative and qualitative research
RA Qualifications: Undergraduate with an interest in public policy, economics, and related topics.
Understanding unconditional cash transfers in the United States
Faculty Mentor: Adrienne Sabety
We are creating cash transfer studies to understand the impact of unconditional cash on individuals outcomes
RA Responsibilities: Coding and understanding the literature
RA Qualifications: Basic to advanced coding skills.
Unbiased Covariate Adjustments in Clustered Experiments
Faculty Mentor: Jann Spiess
We are interested in reanalyzing data from existing experimental research using new data analysis and machine-learning tools that we've developed to make better use of available data. In particular, we're interested in reanalyzing clustered experiments, which, unlike traditional experiments, typically have very few experimental units (e.g., villages or hospitals). Despite having limited experimental units, clustered experiments often collect detailed information about a much larger number of individuals (e.g., villagers or doctors). Our belief is that by making better use of all of this rich data on individuals, we can get more precise estimates without introducing any systematic bias.
RA Responsibilities: The RA will be responsible for systematically collecting data from existing clustered experiments and reanalyzing them. The data collection process will include creating a catalog of existing experiments and acquiring their data when feasible. The analysis of the experiments will include the replication of the original authors' results as well as the application of our new method to reanalyze the data. The implementation of our method will entail applying various machine-learning techniques that the RA may have learned about in their coursework.
RA Qualifications: Some coding experience (e.g., R, Python, Matlab or Julia); Interest in machine learning techniques; Some prior exposure to statistics or data analysis (e.g. an econometrics class)
Three Centuries of International Capital Flows
Faculty Mentor: Chenzi Xu
Foreign capital flows are an important source of funding for emerging markets with less-developed domestic financial markets, but foreign flows often come with risks: through them, countries can "import" credit market fluctuations from the global financial center. While foreign flows have been recognized as a source of disruption in recent financial crises such as the Asian crises of the 1990s and the global financial crisis in 2008, they also have a long historical precedent. A primary goal of this project is to construct a comprehensive database of granular short-term capital flows around the world to answer a variety of questions related to international finance and macroeconomics. The main focus is to work on constructing two novel historical datasets: one based on the international holdings of the first global bond market in London, and one based on the short-term capital flows of global banks around the world.
RA Responsibilities: coding, data cleaning, reading
RA Qualifications: stata/python/R or similar (or willingness to learn); or experience with history (classes, RA work)