Leveraging Natural Language Processing Techniques for the Analysis of Open-ended Survey Responses in Public Opinion Research |
|
| Coordinator 1 | Professor Peter Selb (University of Konstanz) |
Open-ended survey questions allow respondents to express their views freely, unconstrained by predetermined options. This approach captures unexpected or novel responses and yields rich, contextual data that can illuminate underlying attitudes and motivations. However, the time-consuming nature of manually coding open-ended responses has historically limited their use, especially in large-scale studies.
This session explores how recent devvelopments in NLP techniques, including large language models, are empowering researchers to tap into the full potential of open-ended survey questions. We welcome both methodological contributions and case studies that showcase the application of these techniques in public opinion surveys. Special emphasis will be placed on attitude measurement.