In a world of changing research practices, three goals for science and scholarship can be identified: making research more open, efficient and ‘good’. The latter category involves research integrity, fair credit, transparency and reproducibility, and is potentially the most transformative for the way research currently is conducted. Open sharing in all phases of the research cycle is necessary, but not sufficient to translate this goal into practice.
Through our global survey and multiple workshops, we have some indications on the extent to which researchers currently identify integrity issues as important, and what practices regarding research integrity they feel should be included in a truly open science workflow. We will also discuss some practical examples of research tools that have integrity checks built in or that contribute to integrity and help avoid questionable research practices.
The extent to which researchers feel enabled or constrained in adopting tools and practices that promote integrity is influenced by many factors, including tool interoperability, internal and external collaboration culture, and assessment and evaluation criteria. At all these levels, stakeholders in scholarly communication (from researchers themselves, publishers, librarians, and tool developers to research institutions, governments and funders, and to some extent even journalists and the general public) can work together to effect a change in research culture.
This should not be mere policing (though Retractionwatch, Pubpeer and work of editors is important, the first two also to create awareness), but checks built into the system itself and (slowly) changing the culture towards open ‘kitchens’, acceptance of failure, null/negative results pubs etc.
The current mainstream journal publication system, with the incentives attached to it, forms part of the problem. There we should move towards a culture of research-led, not publisher-led, scholarly communication.