Computational Social Science for Sustainability (EBS 181/281)
Winter 2025
Schedule and location
Monday lectures and Wednesday lab sections 11:30 AM - 12:50 PM in 320-109, Winter quarter 2025.
Instructor
Dr. Matthew A. Turner, PhD (email)
Feel free to call me by my first name.
Please see my professional site to learn more about me.
Office hours
- Monday and Tuesday 1 - 3 PM (Y2E2 352)
- Thursday 3 - 4 PM (Zoom; link to be shared on Canvas)
- By appointment
Getting help
I recommend seeking help early if students feel stuck or lost. Please attend office hours, email me to ask questions or set up a time to meet, or seek the help from other students.
Accessibility
Students who need an academic accommodation based on the impact of a disability must initiate the request with the Office of Accessible Education (OAE). Professional staff will evaluate the request with required documentation, recommend reasonable accommodations, and prepare an Accommodation Letter for faculty. Unless the student has a temporary disability, Accommodation letters are issued for the entire academic year. Students should contact the OAE as soon as possible since timely notice is needed to coordinate accommodations. The OAE is located at 563 Salvatierra Walk (phone: 650-723-1066).
Course overview
The development and diffusion of sustainable innovations, cooperation for sustainable resource management, and political polarization that can undermine these, share a common explanation: these phenomena all emerge from repeated social interactions between individuals over time. These interactions take the form of social learning, social influence, or strategic economic cooperation.
Take this course to learn how to develop computational “experiments” of repeated social interaction that can be used to design more effective sustainability interventions and to analyze behavioral data. Students will learn transferrable technical skills in programming and mathematics. Students will deepen their interdisciplinary understanding of the social and behavioral sciences. With these skills and understanding, students will be empowered to create research products that analyze and evaluate potential sustainability interventions using computation.
Learning goals
Through participation in lecture and lab sections and completion of course activities, students will
- learn modern approaches to scientific modeling and statistical analysis of social behavior
- learn to develop, implement, and analyze their own models for designing sustainability interventions
- learn to write effective research papers using the Introduction, Model, Analysis, Discussion structure
- improve their R programming skills, including the popular
tidyverse, for simulation modeling and data science see the R for Data Science book for more details.
Expectations
Understanding mutual expectations can help everyone succeed. There are things you can expect of me and things I expect of enrolled students.
What you can expect of me
I will do my best to promote an encouraging, safe, and fair learning environment to promote student success. I will strive to understand and support student career goals coming from a diversity of life experiences.
You can expect I will be eager to help when needed, especially if you start from very little to no experience with math and programming. I understand that these subjects can cause anxiety for some.
Expectations of all students
Students are expected attend all scheduled course meetings unless there are extenuating circumstances. Please email if this occurs. Students are expected to seek help if they are struggling or stuck.
Expectations of graduate students
Graduate students will be required to complete an extra exercise on each problem set that will be extra credit for undergraduates. They will be held to higher standards for clarity, structure, and technical detail in midterm and final projects.
Honor code
You are expected to cite sources and individuals from whom you have learned and borrowed ideas as a display of academic, intellectual, and creative integrity. Failure to do so is a violation of Stanford’s Honor Code and is a serious offense, even when the violation is unintentional. Conduct prohibited by the Honor Code includes all forms of academic dishonesty, among them unpermitted collaboration and representing others’ work as one’s own. Please review the policies and guidance from the Office of Community Standards, and documentation and citation resources from the Hume Center for Writing and Speaking.
Course materials
Students will need a laptop or otherwise portable computer to bring to the Wednesday lab sections. There are a number of readings from journals and books (see the Calendar below), but these are either available through Stanford Libraries, or if not I will provide PDF copies via Canvas.
Course structure
Each week will have a Monday lecture on topics in computational social science for sustainability. Wednesday meetings will focus on developing programming, analysis, and writing skills in an interactive lab-section setting. In these Wednesday sections students will be introduced to problem sets and midterm and final projects, and have time to work together with the instructor and peers.
Coursework and Grading
Students will be evaluated based on their completion of six assignments worth 100 points total: four problem sets (10 points each), a midterm project (20 points), and a final project (40 points). Undergraduate students will have the opportunity for bonus points on each problem set.
- Problem sets: There will be four problem sets introduced on Wednesdays during the computing lab section. Students will also have the opportunity to work together and ask the instructor questions during other lab sections before each assignment is due. (10 points per problem set)
- Midterm project: Students will write a report on how they will use a model from the course to address a sustainability problem of interest. The midterm project will be used as a foundation for the final project. (20 points)
- Final project: Students will expand on their midterm project, performing a detailed model analysis and discussing the implications of their results for designing sustainability interventions. (40 points)
Late coursework policy
If there is a family or health emergency or other acute distress please contact me to make arrangements to submit late work without penalty. Otherwise the following policy applies:
- Problem sets up to 72 hours late can receive 50% credit.
- Problem sets up to one week late can receive 20% credit.
- The midterm project may be submitted up to one week late to receive 50% credit.
- The final project may not be submitted late.
- No credit will be given for work beyond one week late.
Lecture and lab calendar and topics outline
Calendar
In the calendar below, PS stands for problem set. Subject to change.
| Week | Topic | Coursework | Readings |
|---|---|---|---|
| 1, M 1/6 and W 1/8 | What is social science, why computation, and how these can promote sustainability in socio-ecological systems. |
|
|
| 2, M 1/13 and W 1/15 | The effect of social networks on sustainable innovation development and diffusion. | ||
| 3, M 1/20 and W 1/22 | How asymmetric preferences for within-group interaction can create sustainability-promoting social networks, and how to measure this in the real world. |
|
|
| 4, W 1/29 (no class M 1/27) | Social influence represented as forces causing opinion dynamics. | ||
| 5, M 2/3 and W 2/5 | Opinion dynamics in the context of sustainability. Experimental design and measurement in opinion dynamics experiments. |
|
|
| 6, M 2/10 and W 2/12 | Common-pool resource management dilemmas: when and why do people cooperate? (Part I) |
|
|
| 7, W 2/19 (no class M 2/17) | Common-pool resource management dilemmas: when and why do people cooperate? (Part II) |
|
|
| 8, M 2/24 and W 2/26 | How to perform and report computational social model analyses. |
|
|
| 9, M 3/3 and W 3/5 | Institutions support cooperation by balancing variation and homogeneity within and between stakeholder groups. | ||
| 10, M 3/10 and W 3/12 | Review: A look back at how computational social science can promote sustainability, through the lens of the Price equation. |
|
Course outline
Computational social science can help design sustainability interventions. Social science theory provides models of repeated human interaction over time that can be used, for example, to represent Ostrom’s eight “design principles” for sustainable socio-ecological systems.
- Lab: Introducing the 80% success rate exercise, “How much advertising is necessary for an 80% success rate in spreading a sustainable innovation in groups, given population size and average number of acquaintances of people in the group?”
How human psychology, groups, and social networks can promote or inhibit the diffusion of sustainable innovations, Part I: single-group social networks.
- Lab: Could innovation-supporting social networks also promote inequality (Moser and Smaldino 2023)?
How human psychology, groups, and social networks can promote or inhibit the diffusion of sustainable innovations, Part II: two-group (or more) social networks, i.e., metapopulation social networks.
Lab I: 80% success rate exercise with two-group social networks defined by each group’s homophily level, i.e., tendency of group members to interact with others from their own group (Matthew A. Turner et al. 2023).
Lab II: Use stochastic block model to infer networks from data (De Bacco et al. 2023; Ross, McElreath, and Redhead 2024).
Social influence: understanding the effect of rhetoric as a force that acts on opinions and beliefs. How to measure opinion dynamics and
- Lab I: When is polarization path-dependent and therefore possible to avoid (Matthew A. Turner and Smaldino 2018)?
- Lab II: Opinion dynamics measurement depends on accurate inference using categorical (Likert-style) observational data (Liddell and Kruschke 2018).
The emergence of cooperation via reciprocity: application to groundwater sustainability. How to predict and restrict potential free-riding based on marginal utility in managing common pool resources using game theory (see Jackson (2008)).
- Lab I: “Groundwater sharing dilemma” (as we’ll call it, though it’s just a re-telling of the famous prisoners’ dilemma)
- Lab II: Agent-based model of behavioral study of “avoidance of disastrous climate change in a public goods game” by Tavoni et al. (2011)
Ideal institutions support human cooperation by balancing variation and maintenance of beneficial behaviors within and between stakeholder groups (Richerson et al. 2016; Waring et al. 2015). Example: sustainable agricultural practices like crop switching (Waring et al. 2023; Kling et al. 2024).
- Lab I: The evolution of property rights supports sustainability (Waring, Goff, and Smaldino 2017). What sorts of social networks evolve? Could alternatives better promote or inhibit the development and diffusion of innovations?