Machine Learning Research Assistant for Behavioral Science and Finance Studies
Diag Davenport, Professor
Information, School of
Applications for Fall 2025 are closed for this project.
Project Overview
This research bridges machine learning with behavioral science, finance, and law to study how algorithms can model human interaction and predict institutional outcomes.
Illustrative Studies
Team Performance: Train ML models on recordings of group collaboration to identify behavioral patterns predictive of team success.
Startup Success: Use profiles of ventures—team traits, market position, funding—to predict key success indicators in venture capital.
Legal Applications: Extend methods to law, e.g. assessing grand jury outcomes and detecting bias in local laws.
RA Role
The RA will support data processing, literature review, and logistics across projects, contributing to papers and presentations in behavioral science, finance, and legal analytics.
Role: Role for the Undergraduate Research Assistant:
The undergraduate research assistant (RA) will play a hands-on role in supporting both the behavioral science and machine learning components of this project. Working closely with a post-doctoral fellow and the faculty advisor, the RA will gain practical machine learning experience while developing key research and organizational skills.
Core Tasks:
Data Handling and Preparation:
Assist in cleaning, organizing, and processing raw data from both team interaction recordings and startup profile datasets.
Work on feature engineering tasks to prepare data for supervised learning models, identifying meaningful variables that contribute to predictions in both team dynamics and startup success.
Manage and catalog data to ensure accuracy, consistency, and reproducibility.
Model Development and Analysis:
Collaborate on the development of machine learning models for each project, focusing on supervised learning techniques.
Conduct exploratory data analysis (EDA) to identify trends, outliers, and key insights within the datasets.
Support model training, validation, and tuning, learning how different modeling decisions impact predictive performance.
Literature Review and Background Research:
Conduct a comprehensive literature review on relevant behavioral science and finance topics to inform research questions and model interpretations.
Summarize and organize findings into concise, accessible documents that support the research team’s writing and presentations.
Administrative and Logistical Support:
Assist with logistics for survey administration and participant management, including proofreading materials, checking survey settings, and managing data collection timelines.
Prepare and submit abstracts, research summaries, and presentations for conferences.
Create presentation slides for internal updates and external conference presentations, ensuring clarity and professionalism.
Learning Outcomes:
Machine Learning Skills: The RA will gain foundational skills in supervised machine learning, including model selection, feature engineering, and model evaluation, within real-world applications in psychology and finance.
Data Analysis Proficiency: By working with large and diverse datasets, the RA will improve their ability to clean, manipulate, and interpret data, gaining experience with exploratory data analysis (EDA) and learning about the challenges of handling behavioral and economic data.
Research and Writing Skills: Through literature reviews and administrative support for research papers, the RA will enhance their ability to synthesize academic literature and contribute to empirical writing. This experience will deepen their understanding of research communication and academic publication processes.
Presentation and Communication: Preparing slides and supporting conference applications will build the RA’s presentation skills, helping them learn how to communicate complex research findings in accessible and visually compelling ways.
This role is designed to provide a comprehensive experience in empirical research, helping the RA build a robust skill set that bridges machine learning, behavioral science, and finance.
Qualifications: Technical Skills:
Machine Learning: Basic knowledge of supervised learning models and evaluation metrics.
Programming: Experience with Python (preferred) or R, including ML libraries like scikit-learn or TensorFlow.
Data Management: Familiarity with data cleaning and feature engineering.
Quantitative and Analytical Skills:
Statistics: Understanding of basic statistics (e.g., descriptive stats, correlations) and, ideally, regression.
Attention to Detail: Accuracy in data handling and result-checking.
Research and Writing Skills:
Literature Review: Ability to summarize research insights.
Academic Writing: Proficiency in summarizing, proofreading, and drafting submissions.
Organizational and Communication Skills:
Effective Communication: Clear presentation and writing.
Time Management: Ability to meet deadlines and manage tasks.
Teamwork: Ability to collaborate and receive direction.
Relevant Background:
Coursework: Introductory classes in machine learning, data science, or behavioral science.
Interest in Research: Enthusiasm for applying ML to social sciences.
Day-to-day supervisor for this project: Denis Peskoff, Post-Doc
Hours: 12 or more hours
Education, Cognition & Psychology Social Sciences