Dataartikelns roll i vetenskaplig kommunikation : En intervjustudie med forskare vid svenska universitet

Sammanfattning: To make not only scholarly publications but also the data underlying those publications openly accessible is part of the movement for Open Science. There are several arguments for sharing research data, for example to promote science and collaboration, to make validation of results possible and to reduce redundant data production (Callaghan et al. 2012; Tenopir et al. 2011). Open Data policies are a more and more common strategy from governments, research funders as well as universities. Still – a lot of data is never shared and several studies have pointed out a number of reasons for this: Researchers don’t feel they get properly rewarded for making data useful for others, they lack time and they fear that their research might get scooped. There are also ethical concerns around sensitive data and questions around intellectual property (Kratz & Strasser 2015; Tenopir et al. 2011). Sometimes researchers just don’t consider their data to be relevant for others (Tam 2016). The data paper is ”a journal publication whose primary purpose is to describe data, rather than to report a research investigation. […] Its purposes are threefold: to provide a citable journal publication that brings scholarly credit to data publishers; to describe the data in a structured human-readable form; and to bring the existence of the data to the attention of the scholarly community” (Chavan & Penev 2011, s. 3). While the reasons for not sharing data have been studied thourougly (e.g.Van den Eynden & Bishop 2014) less is known about the incentives for researchers to actually share their data, and even less about what role the data paper can have. The aim of this study therefore is to investigate what functions a data paper can have for researchers data sharing pracitices, how it differs from other ways of sharing and reusing data and what significance it has that a data paper can be reviewed, cited and a part of the academic merit system. The theoretical framework is based on three different perspectives: Epistemical cultures (Knorr Cetina 1999), the socio-technical perspective (Bijker 1995; Van House 2003a) and the document perspective (Frohmann 2004). Qualitative interviews were performed with eight researchers who work in different research areas and in different positons at Swedish universities. All of them have written and/or cited data papers or otherwise shared their data actively Some of my findings are that: writing a data paper can be a faster way of publishing data, since you don't have to write a conventional article with hypotheses and argumentation; in a data paper you can publish nul and negative findings; the data paper makes it easier to cite data in a formalized way; and it saves space in the applied article when you can describe data in detail in the data paper. Also the interviewed researchers see no risk that their openess will be misused and the fact that a data paper can be added to your cv is regarded as something that can be beneficiary for future career and possibilities of recieveing funding. The review process can enhance the data and gives it a quality mark.