A cross-sectional audit and survey of Open Science and Data Sharing practices at The Montreal Neurological Institute-Hospital

Abstract Objectives To audit all publications produced by Montreal Neurological Institute-Hospital researchers regarding open science practices and to survey Neuro-based researchers about barriers and facilitators to data sharing. Setting, design and participants In the first study, we retrieved 313 unique publications and collated all Neuro publications from 2019 and extracted information from each article pertaining to data sharing and other open science practices. We included all empirical papers and pre-prints that were reported in English. In the second study, one hundred twenty-four participants (out of 553) completed the survey, a response rate of 22.42%. We surveyed all Neuro researchers. Primary and secondary outcomes for the audit we examined data sharing and open science practices. For the survey, we asked participants about their data sharing practices. Results We found that 66.5% of these publications (n=208) included a data sharing statement. Overall, 74.5% (n=155) of articles had data that was publicly available. When examining broader open science practices, rates of compliance tended to be lower. For example, 94.9% (n=297) of publications failed to register a protocol. Among participants who had published a first or last authored paper in the past year, most participants, 53 of 74 (71.62%), reported that they had openly shared their research data. Less than half of the participants 37.50% (n=45) reported having engaged in training related to data sharing within the last 12 months. Conclusion We found that half of all publications included in the audit shared data. Participants indicated an appetite for resources for learning about data sharing signaling a willingness to perform better. Strengths and limitations of this study To serve as a baseline to benchmark for improvements in data sharing and other open science practices To measure progress over time. The results of the study cannot be generalized. It is hard to measure changes in the community.