Elamur Research⌕
☰
home / papers / 2026 / Question and Response Dynamics in Public Service Encounters

Question and Response Dynamics in Public Service Encounters

Summary

Face-to-face meetings between citizens and street-level bureaucrats, such as those in job centers and welfare offices, are where officials exercise much of their discretion. Yet recordings of these meetings are rare because privacy concerns make them hard to obtain. The authors release PSE v2.0, a German corpus of 232 anonymized verbatim transcripts. It roughly doubles the earlier PSE v1.0 (106 conversations) and adds tax, registration and social welfare offices to the original job-center focus. Collection and anonymization follow the same protocol as v1.0. The authors also extract all 12,355 question–response sequences. Questions are labelled by syntactic form and responses by pragmatic function. To pick an automatic question-type annotator, they compared the rule-based LiAnS system with GPT-5 on an expert-labelled sample. GPT-5 scored higher and was used for the whole corpus. The resource supports quantitative research on how officials steer conversations through questions and how citizens answer, deflect or push back. It also offers enough data to study less frequent phenomena that the smaller first release could not.

Key points

  • Corpus growth (PSE v1.0 → PSE v2.0):
    • Conversations: 106 → 232
    • Tokens: 433,780 → 740,973
    • Utterances: about 57,119 in v2.0
    • Questions: 6,033 → 12,355
  • Questions and interruptions: About 19.6% of utterances contain at least one question, and 37.5% contain an interruption. Of the questions, 11,833 are complete and 522 are interrupted; the interrupted ones were reconstructed and included.
  • Question scheme: Five syntactic types: Polar, Tag, Alternative, Wh and Other. A gold sample of 747 questions (about 6%) was annotated by one expert linguist. On this sample, GPT-5 reached about 0.75 accuracy, compared with about 0.65 for LiAnS.
  • Response scheme: Six categories cover cases where no reply follows, where the reply is cut off by an interruption, where the reply comes as a counter-question, where the reply evades the question (a non-answer), where a proper answer is given, and an Other category. Responses were labelled with an ordered, rule-like GPT-5 prompt.
  • Availability: Released under CC-BY 4.0 on Harvard Dataverse (doi:10.7910/DVN/Y8YU2O) in CSV, TXT and XML formats. PSE v1.0 had been downloaded more than 4,000 times.

Our summary of the paper, in our own words — not its text. The work belongs to its authors; read it at the links above.

Fields: Linguistics & languages

Cite

The paper itself, not this page. BibTeX from the library's record of the work. DOI 10.63317/44ysuy55vyoi

@inproceedings{siskou2026question,
  title = {{Question and Response Dynamics in Public Service Encounters}},
  author = {Siskou, Wassiliki and Espinoza, Ingrid and Friedrich, Laurin and Eckhard, Steffen and Hautli-Janisz, Annette},
  booktitle = {Proceedings of the Fifteenth Language Resources and Evaluation Conference},
  year = {2026},
  doi = {10.63317/44ysuy55vyoi},
  url = {https://doi.org/10.63317/44ysuy55vyoi}
}

References

29 references, 2 of them with a card of ours. ↗ leads to the work, «find» searches for one we have no id for yet.

  1. Ginger Berninger and Catherine Garvey. 1981. Relevant replies to questions: Answers versus evasions. Journal of Psycholinguistic Research, 10(4):403–420. find
  2. Anders Bruhn and Mats Ekström. 2017. Towards a Multi-Level Approach on Frontline Interactions in the Public Sector: Institutional Transformations and the Dynamics of Real-time Interactions. Social and Policy Administration, 51(1). find
  3. Sasha Calhoun, Jean Carletta, Jason M. Brenier, Neil Mayo, Dan Jurafsky, Mark Steedman, and David Beaver. 2010. The nxt-format switchboard corpus: a rich resource for investigating the syntax, semantics, pragmatics and prosody of dialogue. Language Resources and Evaluation, 44(4):387–419. find
  4. Jean Carletta, Amy Isard, Stephen Isard, Jacqueline C. Kowtko, Gwyneth Doherty-Sneddon, and Anne H. Anderson. 1997. The reliability of a dialogue structure coding scheme. Computational Linguistics, 23(1):13–31. find our card
  5. Steffen Eckhard and Laurin Friedrich. 2022. Linguistic Features of Public Service Encounters: How Spoken Administrative Language Affects Citizen Satisfaction. Journal of Public Administration Research and Theory. find
  6. Steffen Eckhard and Laurin Friedrich. 2025. The language of public encounters: Computational measures of complexity and emotionality in spoken bureaucratic communication. Social Policy & Administration. find
  7. Steffen Eckhard, Laurin Friedrich, Annette Hautli-Janisz, Vanessa Mueden, and Ingrid Espinoza. 2022. A Taxonomy of Administrative Language in Public Service Encounters. International Public Management Journal. find
  8. Ingrid Espinoza, Steffen Frenzel, Laurin Friedrich, Wassiliki Siskou, Steffen Eckhard, and Annette Hautli-Janisz. 2024. PSE v1.0: The first open access corpus of public service encounters. In Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and… find
  9. Steffen Frenzel and Annette Hautli-Janisz. 2025. Identifying small talk in natural conversations. In Proceedings of the 9th Joint SIGHUM Workshop on Computational Linguistics for Cultural Heritage, Social Sciences, Humanities and Literature (LaTeCH-CLfL 2025), pages 272–277, Albuquerque, New Mexico.… find
  10. Laurin Friedrich and Steffen Eckhard. 2025. Inequality in frontline communication: bureaucrats talk differently to men and women. Journal of Public Administration Research and Theory. find
  11. Fabrizio Gilardi, Meysam Alizadeh, and Maël Kubli. 2023. ChatGPT outperforms crowd workers for text-annotation tasks. Proceedings of the National Academy of Sciences, 120(30):e2305016120. find our card
  12. Jonathan Ginzburg, Zulipiye Yusupujiang, Chuyuan Li, Kexin Ren, and Paweł Łupkowski. 2019. Characterizing the response space of questions: a corpus study for English and Polish. In Proceedings of the 20th Annual SIGdial Meeting on Discourse and Dialogue, pages 320–330, Stockholm, Sweden. Association… find
  13. Valentin Gold, Mennatallah El-Assady, Annette Hautli-Janisz, Tina Bögel, Christian Rohrdantz, Miriam Butt, Katharina Holzinger, and Daniel Keim. 2015. Visual linguistic analysis of political discussions: Measuring deliberative quality. Digital Scholarship in the Humanities, 32(1):141–158. _eprint: h… find
  14. Annette Hautli-Janisz, Katarzyna Budzynska, Conor McKillop, Brian Plüss, Valentin Gold, and Chris Reed. 2022. Questions in argumentative dialogue. Journal of Pragmatics, 188:56–79. find
  15. Peter Hupe and Michael Hill. 2007. Street-Level Bureaucracy and Public Accountability. Public Administration, 85(2). find
  16. Daniel Jurafsky, Elizabeth Shriberg, and Debra Biasca. 1997. Switchboard swbd-damsl shallow-discourse-function annotation – coders manual, draft 13. Technical report, University of Colorado Institute of Cognitive Science. find
  17. Michael Lipsky. 2010. Street-Level Bureaucracy: Dilemmas of the Individual in Public Service. American Political Science Association, 76(1). find
  18. Steven Maynard-Moody and Michael Musheno. 2000. State Agent or Citizen Agent: Two Narratives of Discretion. Journal of Public Administration Research and Theory: J-PART, 10(2). find
  19. Gaël Le Mens, Balázs Kovács, Michael T. Hannan, and Guillem Pros. 2023. Uncovering the semantics of concepts using GPT-4. Proceedings of the National Academy of Sciences, 120(49):e2309350120. find
  20. Nadine Raaphorst. 2021. Administrative Justice in Street-Level Decision Making: Equal Treatment and Responsiveness. In The Oxford Handbook of Administrative Justice, pages 1–30. Oxford University Press. find
  21. Wassiliki Siskou and Ingrid Espinoza. 2024. “So, are you a different person today?” Analyzing Bias in Questions during Parole Hearings. In Proceedings of the Second Workshop on Social Influence in Conversations (SICon 2024), pages 116–128, Miami, Florida, USA. Association for Computational Linguisti… find
  22. Wassiliki Siskou, Laurin Friedrich, Steffen Eckhard, Ingrid Espinoza, and Annette Hautli-Janisz. 2022. Measuring plain language in public service encounters. In Proceedings of the 2nd Workshop on Computational Linguistics for Political Text Analysis (CPSS-2022) Potsdam, Germany. find
  23. Tanya Stivers and N.J. Enfield. 2010. A coding scheme for question-response sequences in conversation. Journal of Pragmatics, 42(10):2620 – 2626. Question-Response Sequences in Conversation across Ten Languages. find
  24. Kalsoom Sumra. 2019. Social Equity in Public Administration: Fairness, Justice and Equity, tools for social change. Pakistan Administrative Review, 3(1):1–15. find
  25. Konstantinos Thomas, Giorgos Filandrianos, Maria Lymperaiou, Chrysoula Zerva, and Giorgos Stamou. 2024. “I never said that”: A dataset, taxonomy and baselines on response clarity classification. In Findings of the Association for Computational Linguistics: EMNLP 2024, pages 5204– 5233, Miami, Florid… find
  26. Joanna Thornborrow. 2002. Power talk : language and interaction in institutional discourse. Real language series. Longman. find
  27. Bernardo Zacka. 2019. Street-Level Bureaucracy and Democratic Theory. In Research Handbook on Street-Level Bureaucracy, pages 448–461. Edward Elgar Publishing. find
  28. Paweł Łupkowski and Jonathan Ginzburg. 2017. Query responses. Journal of Language Modelling, 4(2):245–292. 8. find
  29. Language Resource References Ingrid Espinoza, Steffen Frenzel, Laurin Friedrich, Wassiliki Siskou, Steffen Eckhard, and Annette Hautli-Janisz. 2024. PSE v1.0: The First Open Access Corpus of Public Service Encounters. In Proceedings of the 2024 Joint International Conference on Computational Linguis… find

↑ back to top

This page in Markdown → https://md.research.elamur.ai/free/en/papers/siskou-2026-question-response-dynamics-public-service/