Research question/goal:

Social science research increasingly draws on “digital behaviour” data, such as browsing histories, to complement traditional survey data. These data can be used, for example, to capture participants’ political news consumption online in addition to their responses to political questions in surveys. However, past research has not yet fully exploited the rich potential of browser history data but rather extracted ad-hoc measures to infer what political messages participants receive. Meanwhile, large language models (LLMs) have proven effective in a range of data processing tasks and offer promising opportunities to capture multi-modal content (e.g., text, images, and videos on websites). 

This project adapts LLMs to analyse and condense browsing histories with the aim of gaining deeper insights into online behaviour. The project thereby opens a new perspective on the relationship between media consumption and political attitudes.