Teaching
Overview
I teach courses on social network analysis, applied statistics and causal inference, program evaluation, and collaborative governance. My primary aim in these courses is to give students the methodological tools to answer questions they actually care about, and the judgment to know when a method fits the question and when it does not. My courses are taught in R and Quarto, and students leave with reproducible workflows.
My teaching has been recognized with UIC’s Teaching Recognition Award (2016 and 2021), the Silver Circle Teaching Award (2018), and the Honoring Our Professors’ Excellence (HOPE) Award (2017).
Courses below are those I have taught recently, each with a link to the most recent syllabus.
Doctoral Courses
Advanced Data Analysis I
PA 541 Syllabus
Modern regression and causal inference for applied researchers. Covers ordinary least squares and its assumptions, interactions and moderation, log transformations and functional form, generalized linear models and logistic regression, maximum likelihood estimation, panel data and fixed effects, and difference-in-differences. Emphasis throughout is on interpretation and on directed acyclic graphs as a tool for reasoning about what a model can and cannot identify. Taught in R and Quarto.
Advanced Data Analysis II
PA 542 Syllabus
A second course in applied methods, organized around three sections: the linear model and its extensions, causal inference methods, and multilevel models. Topics include count regression models, regression discontinuity designs, instrumental variables, matching methods, and multilevel and longitudinal multilevel models, with emphasis throughout on conceptual understanding and on matching a method to one’s data, sampling strategy, and theory.
Collaborative Management and Governance Networks
PA 513 Syllabus
Contracting out, devolution, and the growing complexity of policy problems have pushed more and more public work into networks of agencies, nonprofits, firms, and citizen groups. This doctoral seminar takes up the hard questions that follow: are networks actually more effective than hierarchies, how should performance be measured across organizations that define success differently, and what brings these arrangements together or holds them apart? The course works through network types and governance modes, theories of formation and collaborative governance, learning and knowledge transfer, network management and leadership, performance and effectiveness, and the tension between network governance and democratic accountability — what Sørensen and Torfing call democratic anchorage. It closes on street-level and intra-organizational networks, where policy actually gets implemented.
Masters Courses
Program Evaluation
PA 528 Syllabus
Impact evaluation and causal inference for public programs. The course is built around a practical question — did this program produce the results it claimed? — and works through the designs that can credibly answer it: randomized experiments, matching, difference-in-differences, regression discontinuity, and instrumental variables. Also covers logic models and theory of change, directed acyclic graphs, threats to validity, and the ethics and limits of evaluation evidence. Taught in R and Quarto.
Doctoral Advising
I serve as Director of Graduate Studies for the PhD program in Public Administration and chair or serve on dissertation committees in public administration, urban planning, education, and engineering. I am generally interested in working with students on interorganizational networks, collaborative governance, and network methods. Prospective students are welcome to reach out.
Social Network Analysis
PA 594 Syllabus
We begin with network-based research design and the practical work of collecting and managing relational data, including the ethical questions that arise when respondents can be identified by their ties. From there the course moves through visualization, descriptive analysis of ego and whole networks, centrality, structural holes and social capital, and then to inferential methods: MRQAP and network regression, exponential random graph models, and stochastic actor-oriented models for longitudinal and coevolution data.