:partying_face: Full Paper accepted in PLoS One! :book:

Our paper on using GPT-4 to tackle the cold start problem in adaptive political surveys (Bachmann et al., 2025) has been accepted in PLOS ONE!

Adaptive questionnaires dynamically select questions based on previous answers, but they need training data to work well. We explored whether Large Language Models can generate synthetic user interactions to pre-train these systems before real data is available.

Using data from the Swiss Voting Advice Application Smartvote, we showed that GPT-4 can accurately simulate party positions and that pre-training with synthetic data significantly improves recommendation accuracy compared to random initialization.

Thank you to all my co-authors (Daan van der Weijden, Lucien Heitz, Cristina Sarasua, and Abraham Bernstein), the anonymous reviewers, and the academic editor!

Adaptive questionnaires dynamically select the next question for a survey participant based on their previous answers. Due to digitalisation, they have become a viable alternative to traditional surveys in application areas such as political science. One limitation, however, is their dependency on data to train the model for question selection. Often, such training data (i.e., user interactions) are unavailable a priori. To address this problem, we (i) test whether Large Language Models (LLM) can accurately generate such interaction data and (ii) explore if these synthetic data can be used to pre-train the statistical model of an adaptive political survey. To evaluate this approach, we utilise existing data from the Swiss Voting Advice Application (VAA) Smartvote in two ways: First, we compare the distribution of LLM-generated synthetic data to the real distribution to assess its similarity. Second, we compare the performance of an adaptive questionnaire that is randomly initialised with one pre-trained on synthetic data to assess their suitability for training. We benchmark these results against an “oracle” questionnaire with perfect prior knowledge. We find that an off-the-shelf LLM (GPT-4) accurately generates answers to the Smartvote questionnaire from the perspective of different Swiss parties. Furthermore, we demonstrate that initialising the statistical model with synthetic data can (i) significantly reduce the error in predicting user responses and (ii) increase the candidate recommendation accuracy of the VAA. Our work emphasises the considerable potential of LLMs to create training data to improve the data collection process in adaptive questionnaires in LLM-affine areas such as political surveys.