Teaching computational social science at UMass Amherst has been an eye-opening experience. UMass stands out for blending data science and social science in a way that few universities do, and my students come from diverse backgrounds—sociology, political science, economics, even the humanities—all eager to explore how coding, big data, and machine learning can illuminate complex social questions.
At first, many students approach the course with hesitation. Some doubt their technical abilities, while others question whether sociology and data science can meaningfully intersect. But as the semester progresses, I’ve watched students transform—not just in their technical capabilities, but in how they approach research and critical thinking. Here’s what I’ve learned from teaching in this space.
One of the things I love about UMass Amherst is that it attracts students who don’t want to be boxed in. They seek more than traditional social science methods, yet they’re not convinced that data science alone can explain human behavior. They want both—theoretical depth and technical skills.
That’s what makes computational social science so exciting here. Students quickly realize that learning to code doesn’t mean abandoning theory—it means gaining the power to test social theories on a massive scale. They arrive thinking they must choose between being a sociologist or a data scientist, and they leave understanding they can be both.
Many of my students come from sociology and political science backgrounds, and most have never written a line of code. Initially, the prospect of learning Python or R terrifies them. They see coding as a domain reserved for engineers and computer scientists.
But then something changes. They write their first script to clean messy survey data. They scrape their first dataset from Twitter. They visualize a social network of misinformation spread. And suddenly, it clicks. They discover that coding isn’t some mystical skill—it’s just another research tool, as accessible as interviews or surveys.
That said, getting over the initial fear is the hardest part. UMass Amherst offers powerful resources—like the Computational Social Science Institute (CSSI) and partnerships with the College of Information and Computer Sciences—that help students overcome this challenge. Once they embrace coding, they create amazing things with it.
UMass maintains a strong tradition of critical thinking in the social sciences. Unlike many data science programs that focus solely on prediction and algorithms, our students learn to ask, ”What does this actually mean?”
I emphasize this point regularly:
Just because an algorithm finds a pattern doesn’t mean that pattern is meaningful.
We explore real-world cases where machine learning models perpetuate bias or where big data leads to misinterpretation. Students develop skills not just in data analysis, but in questioning their results and examining the broader implications.
Computational social science remains a new field, and students occasionally struggle with the lack of a definitive approach. One week we explore network analysis, the next we develop text classifiers, and suddenly a new AI model reshapes our entire perspective.
Initially, this constant evolution overwhelms students. However, they soon discover that adaptability is their most valuable asset. I emphasize these points:
By graduation, students feel empowered—not because they’ve mastered every tool, but because they’ve learned how to keep learning.
One of my greatest joys in teaching here is tracking my students’ success. Some pursue data science roles in policy and research, while others apply computational methods to public health, labor studies, and media analysis. Employers seek out UMass graduates because they possess a rare combination: the ability to bridge social science and data science.
Whether they’re working at tech companies, government agencies, think tanks, or academia, our graduates emerge with a distinctive skill set. They go beyond simply running machine learning models—they think critically about how data affects society and shapes our understanding of social issues.
Teaching computational social science at UMass Amherst has revealed something powerful: this field transcends mere coding skills—it’s fundamentally changing how we study the world. We’re equipping students with new ways to explore inequality, politics, media, and social change, while maintaining strong roots in sociological traditions.
For anyone considering this field—whether as a student, researcher, or educator—just start. Begin with a Python tutorial, dive into a dataset, ask challenging questions, and embrace experimentation. The future of social science needs these new perspectives and approaches.
I’d love to hear from others working at the intersection of data and social science—what’s your experience been like? 🚀
Back