PEER REVIEWED, OPEN DATA PUBLICATION AS MEANS OF DATA QUALITY MANAGEMENT IN RESEARCH

A discussion has begun in recent years, how data can be cited and if or how they could be published while in parallel it was debated whether and why data can or should be openly accessible. The newly launched journal “Earth System Science Data” – a journal, possibly the first, devoted exclusively to peer reviewed publication of datasets – combines some specific answers to these questions in order to address data quality management especially in publicly funded research.

It can be safely assumed that organizations collecting data as essential part of their business, such as exploration companies or public meteorological services, have instituted internal instruments and procedures for data quality management. This is to be expected where financial or other success directly depends on data quality - especially if collected data can be reused and, thus, constitute “capital”.

By contrast, publicly funded research is almost exclusively driven by the requirement to publish (“or perish”). There is no – financial or other - bonus for data quality management as such. Consequently, in most disciplines and institutions it is regarded as overhead - if there is any institutional awareness or policy at all. At the level of an individual researcher, there are more facets to the picture: He will regard collected data as the basis for the next or even many publications, and thus his personal capital. This motivation however will not result in a systematic drive towards neutral quality management practises that must typically involve third parties.

It is this background which led to the belief that data publication via peer reviewed journal articles is a concept which ideally unites a well known and respected instrument and procedure of quality management in research with the necessary incentive, namely to increase the publication count of individuals and institutions.

Not least as a “byproduct” of the specific method of peer-review, namely the two stage open peer review, in practise by a number of high impact journals in earth science topics for some years, it is necessary that the datasets published are openly available to reviewers, commentators and, finally, “readers”. This requirement combines the power of peer review with the purging effect well known from open source.

We will discuss the emerging rules and practises of this journal – for example, authors' and reviewers' guidelines – in some detail and try to predict their effectiveness and limits regarding research data quality management.

Finally, we discuss the role of this journal vis-a-vis data centers or data repositories (and their human actors) and of the relation and technical links between the journals papers and datasets.