Weitzman School of Design, University of Pennsylvania
Department of City Planning
Master of City Planning Program
Master of Urban Spatial Analytics Program
Instuctor: Dr. Elizabeth Delmelle
Creator/Student: Zhanchao Yang
This course teaches advanced spatial analysis and an introduction to data science/machine learning in the urban planning and public policy realm. The class focuses on real-world spatial analysis applications and, in combination with introductory machine learning, provide students with a modern framework for efficiently allocating limited resources across space. Unlike its private sector counterpart, data science in the public or non-profit sector isn't strictly about optimization - it requires the understanding of public goods, governance, and issues of equity. We explore use cases in transportation, housing, public health, land use, criminal justice, and other domains. We will learn novel approaches for understanding and avoiding risks of "algorithmic bias" against communities/people of color as well as communities of different income levels.
This course will generally follow a lecture-lab format to break up the 3-hour block. We will begin with some conceptual and theoretical background on the week’s topic and then transition to hands-on coding practice. Please bring a (charged) laptop to class to participate in the lab section. I would very much appreciate it if you dedicate the brief 3-hour timeslot we have together each week to matters concerning this class alone. Working on assignments for your other courses is a distraction to everyone.
By the end of the semester, students should:
- Understand how to build a predictive model for public policy decision-making applications.
- Effectively evaluate the effectiveness, generalizability, and biases of models.
- Be proficient in the data science workflow – data wrangling, exploration, modeling, and communication.
- Understand how to incorporate spatial variables from various sources into predictive models.
- Steif, Ken “Public Policy Analytics”: https://urbanspatial.github.io/PublicPolicyAnalytics/
- Wickham et al. “R for Data Science”: https://r4ds.hadley.nz/
- Lovelace et al., “Geocomputation with R”: https://r.geocompx.org/
- Walker, Kyle “Analyzing US Census Data: https://walker-data.com/census-r/
- Homework Assignments – There will be 5 homework assignments throughout the semester. Most will be due 2 weeks after they are assigned (except the first one, which is shorter). The work for this course builds on skills taught previously so once you fall behind, it is increasingly difficult to catch up. These will be turned in individually and written up individually.
- Occasional Low-Stakes Quiz – These are to ensure that the core concepts I want you to know, you know. They also help me determine what needs to be taught better. They are very low-stakes, and you receive 1 point for doing it and 0 for not doing it (they must be submitted on time). Because you are not graded on how many you got correct, please don’t cheat (a.d.k. ask AI the answer)!! This is your opportunity to let me know you don’t understand something and for me to then try to explain it better.
- Midterm & Final Projects – There will be an applied, larger midterm and final project. Both will be group-based.