Project Retain is a one-year project led by SPARC Europe as part of The Knowledge Rights 21 (KR21) Programme, funded by Arcadia fund (https://www.arcadiafund.org.uk/) which aims to promote change in European copyright, in this case, to support Open Access. Research on rights retention and open licensing in Europe aims to underpin a campaign to develop copyright policy amongst publishers, funders and institutions to support authors. The results of the project Retain are published in the "Opening Knowledge: Retaining Rights and Licensing in Europe 2023" report - https://doi.org/10.5281/zenodo.8084051. Within the project, the survey was conducted with the aim to understand the policies and support research institutions have put in place in a number of areas affecting their researchers. The survey asks about the following topics: 1. Copyright of scientific publications - how intellectual property that protects original works of authorship applies to the written outputs of research. Specifically: Whether the initial ownership of copyright is with authors or their employing organisation What rights are transfered from authors to publisher 2. Author rights retention - the extent to which researchers retain rather than transferring control of their published research 3. Reuse of scientific publications - ensuring that the written outputs of research can be reused, by the community as a whole but also by the original author in new contexts 4. Open licensing - the promotion of licenses which grant permission to access, re-use and redistribute. The full survey questionnaire file is included in appended to the dataset. The survey was disseminated widely across Europe by SPARC Europe and its partner organisations through newsletters, email lists and social media channels in the autumn of 2022. The survey received 238 unique responses with a completion rate of 71%. A total of 92 responses were excluded from the final analysis because: they provided extremely limited or no detail in their responses they answered ‘no’ when asked if they agreed to take part in the survey their response was identifiably a duplicate A total of 146 unique responses were included in the final dataset and analysed. The deposited dataset is anonymised: the personal data and the names and URLs of the institutions are removed or redacted.