OPEN DATA: NICE PEOPLE CAN’T SHARE! Aurelia Green 1 , Jennifer S. McClung 1 , Nichola J. Raihani 2 , Redouan Bshary 1 , Dominique G. Roche 1,3* BACKGROUND Open data (publicly shared research data) are key in promoting transparency and reproducibility in science. However, open data practices are far from universal and strong normative pressures by funders and publishers are still lacking across disciplines. Even when journals mandate open data, authors often share datasets that are incomplete or difcult for third parties to understand and reuse 1,2,3 . Given the highly competitive nature of academic research, authors might be wary of sharing their data for fear of criticism or of others benefting from their work. As such, ‘scholarly altruism’ is often cited as a key reason for why some authors willingly share their data 4 . CONCLUSION Neither levels of cooperativeness nor psychological motivations were predictive of good data-sharing practices. The fact that highly cooperative researchers share data similarly to their less cooperative colleagues points to an obvious lack of training in data management and sharing practices among scientists. We surveyed researchers’ level of cooperation (i.e. self-reported altruism, response time to a survey, amount of money donated to a charity) and their psychological motivations (i.e. conscientiousness, social desirability, and collectivism) to assess how these traits afect the quality (i.e. completeness and reusability) o f publicly shared datasets associated with 100 papers published in journals that mandate open data. 1 Univtersity of Neuchâtel, Switzerland; 2 University College London, UK; Carleton University, Canada *dominique.roche@mail.mcgill.ca; @dom_roche REFERENCES 1 Roche D, Lanfear R, Binning SA, Haf TM, Schwanz LE, Cain KE, Kokko H, Jennions MD & Kruuk LEB (2014) Troubleshooting public data archiving: Suggestions to increase participation. PLoS Biology 12, e1001779. 2 Roche D, Kruuk L, Lanfear R & Binning S (2015) Public data archiving in ecology and evolution: How well are we doing? PLoS Biology 13, e1002295. 3 Renaut S, Budden A, Gravel D, Poisot T & Peres-Neto P (2018) Management, archiving, and sharing for biologists and the role of research institutions in the technology-oriented age. BioScience 68, 400-411. 4 Kim Y & Stanton J (2015) Institutional and individual factors afecting scientists’ data-sharing behaviours: A multilevel analysis. Journal of the Association for Information Science and Technology 67, 776-799. 1 2 3 4 5 Data completeness score non-respondent 1 2 3 4 5 Data reusability score -1.0 -0.6 -0.2 0 0.2 0.6 1.0 data reusability survey response time charity donation data completeness data reusability survey response time 0.66 0.01 1 0.11 0.17 0.35 FIG. 1 FIG. 2 Completeness and reusability scores for 100 open datasets archived on the repository Dryad by authors who responded (n=52) or did not respond (n=48) to a research questionnaire sent by a graduate student. See Roche et al. (2015) for a detailed description of the scoring scheme. Correlation coefcients (r) between the completeness and reusability of 52 open datasets and three measures of cooperation by researchers. Fixed- efects (gender, seniority, psychological motivations) were controlled for in a Bayesian multivariate model.