Conference Agenda

Data citation is a crucial, but often neglected, component of data curation and data management. This contribution shows how data citation can be moved forward by different stakeholders. The stakeholders included here are data repositories, researchers, and academic journals. For all of them proper data citation practices offer benefits. However, often, data are not cited correctly, implying that the producer of the original data does not get appropriate credit for the work underlying the data production process. Improper data citation also implies that repositories cannot track their impact. In this contribution, we analyse several materials, including survey data and journal policies, and elaborate on the steps that are necessary to improve data citation practices, and the respective roles of repositories, researchers, and journals. We derive lessons learned and recommendations for best practice in data curation work, regarding which we want to engage with the IDCC community. In line with the theme of the IDCC 2026 and with current needs and discussions in data citation, our contribution will also reflect on how AI can help repositories, researchers, and journals in improving data citation practices. The focus of this contribution is on social science data in Switzerland; yet the lessons learned are also relevant for other disciplines and national contexts.

Here we compare the performance and cost of four language models (GPT 4, Llama 3, Gemma 2 and Mixtral 8x7b) in the lightweight task of population group curation. Our findings provide insight into potential sustainable curation practices in the presence of limited resources.