PhD defense of Fynn Bachmann
Congratulations to Fynn Bachmann for defending his PhD thesis “Adaptive Questionnaires: Efficient Preference Elicitation in Digital Democratic Innovations” 🎉
In his dissertation, Fynn Bachmann examines how questionnaires in digital democratic innovations, such as voting advice applications, wiki surveys, and deliberative mini-publics, can be shortened without losing valuable information. His framework predicts participants’ remaining answers and adaptively selects the questions expected to be most informative.
Fynn evaluated the approach through simulations using data from the Swiss voting advice application Smartvote, four online experiments examining how voters respond to adaptive questionnaires, and extensions using generative AI for synthetic training data and conversational surveys. The results will help guide the implementation of adaptive questionnaires by Smartvote for real-world elections starting in 2027.
Fynn was supervised by Prof. Abraham Bernstein as a PhD student in the Dynamic and Distributed Information Systems Group.