Kenneth Joseph
University at Buffalo, NY·Computer Science·Early-career
Social media holds huge amounts of public talk about social issues. It could help explain how Americans form attitudes and how those attitudes vary by place, identity, and social ties.
- Fields
Computer Science
Social Science
Education Research
- Topics
computational social science
social media
natural language processing
- Grants
- 1 known grant, through Aug 2027
What they work on
- Develop mixed methods combining graph-based and natural-language machine learning with qualitative analysis to identify dominant beliefs about social issues in large datasets from X and Parler.
- Collect and analyze a new dataset that links individuals' social media data to their survey responses, revealing where these two measures of attitudes on social issues disagree.
- Study how location, identity, and social network structure relate to people's attitudes by combining measured social media views with data on where users live and whom they connect with.
- Build an agent-based simulation framework grounded in empirical findings and attitude theory to evaluate proposed interventions and designs for online communication systems.
More about the research
Using that data well is hard: online posts may not reflect what people say in surveys, and existing theories of attitude formation were not built for modern online platforms. This project develops computational social science methods that pair large datasets with social theory, and builds tools so researchers without heavy programming training can use them.
Why it matters
Better ways to measure attitudes from social media, checked against survey data, could help researchers and policymakers understand how social differences in views arise. The simulation framework could help designers of online platforms judge the strengths and weaknesses of proposed interventions before rolling them out, and released tools could open these methods to scholars without computational training.
Known grants
From NIH and NSF research grants. Labs often have other funding we can't see.
Closest labs
Filippo Menczer
Indiana University, IN·Computer Science
Build an agent-based simulation of how content spreads through human social networks, modeling individual traits and network structure to predict large-scale information flow.
Nicholas Valentino
University of Michigan, MI·Social Science
Field pre- and post-election interviews for the American National Election Studies and add a 2026 wave to its first-ever ten-year panel, which follows the same respondents over time.
Ted Hsuan Yun Chen
George Mason University, VA·Social Science
Develop a survey-based measurement approach and conceptual framework for belief networks, capturing how attitudes on climate, migration and social fairness influence one another within individuals.
The school
University at Buffalo: 32 labs on PI Pages, $542M a year on research, 316 research doctorates a year.
Reach out
A start for your email. Fill in the brackets.