Sociological Analytics App
“The app allows students and researchers to conduct advanced robust qualitative analysis without needing several years of coding experience.” –Taylor Beauvais College of Arts & Sciences Sociology Department doctoral student
Teaching about cultural differences can be difficult due to the complexity of the subject matter, and attempts to visualize these differences can be hit or miss due to the complicated nature of representation.
Using modern technologies, like web applications, educators can better generate analyses and visual representations of cultural differences and clusters that help solidify concepts of culture to undergraduate and graduate students alike. However, the coding knowledge required to do these analyses and build these three-dimensional visualizations is extensive, and most often, sociology undergraduate students don’t have coding, app development, or advanced statistical knowledge.
To address this problem and leverage emerging educational technologies, Taylor Beauvais, a doctoral student in the College of Arts & Sciences Sociology Department, received Shipley Center Accelerating Classroom Transformation (ACT) funding to build a web app with an interactive dashboard that conducts multiple correspondence analysis (MCA) and creates MCA visualizations using custom qualitative data students provide.
“The research and interview work done for this MCA app project was really interesting and informative, even of patterns that we did not expect to find,” said one student enrolled in Beauvais’ course. “The biggest lesson I learned from this project is to be open to all patterns that the data might show and not just the ones that I intended to research.”
For example, in a unit on how culture and opinions influence voting behavior, students designed a 30-question survey, and uploaded the responses to the app Beauvais developed. From this data set, the app sorted groups into different clusters and found that individuals were less likely to vote if they were technology-heavy users who frequently used X (Twitter) and individuals were more likely to vote for a specific candidate if they used different social media platforms. Ultimately, students learned about the politics of algorithmic authority and how algorithms influence users about candidates.
According to another student enrolled in Beauvais’ course:
“I really loved this project! I think it was cool to combine all the different groups’ survey questions so that if you didn't love the results from your own questions, you had the option to choose any other variables that the MCA analyzed regardless of who was responsible for creating them. Taylor's app was instrumental in that aspect and without him I think the class would have been entirely confused since the software is not easy to pick up.”
Beauvais integrated use of this app in his SO 253 Sociology of Pop Culture and SO 201 Sociological Methods courses throughout 2023 and 2024, impacting over 100 students so far. Pedagogically, this project culminated previous sociology course knowledge through the app to examine and leverage the inter-relationship of information.
“The biggest eye-opening moment for me as an instructor is how we currently teach methods,” said Beauvais. “Often it’s things like a point and click guide on how to use SAS, SPSS, etc., and while it’s useful, I don’t know how good a use of time it is for students to figure out the mechanics of each. This app gives a really good introduction to how these methods work and can be used in students’ own research. It shortens the time frame to learn about analysis and to be able to do it in real-time.”
Moving forward, Beauvais would like to develop a network analysis feature of the app, ultimately making the app a methods toolbox. “After this experience, I now know that if there is a specific method that could be better served by advanced coding, I can build a quick app for that,” said Beauvais.
Project Lead
Taylor Beauvais is a PhD candidate in the sociology department studying algorithms and artificial intelligence. He is a former program lead for BU Spark!’s machine learning and data science classes. He is a graduate fellow at the Rafik B. Hariri Institute for Computing and Computational…
