Data Policies

What is a Data Policy?

The International Federation of Data Organizations for Social Science (IFDO) (n.d.)  provides the following definition of data policies:

Data policies are norms regulating management and publication of research data. They range from recommendations to enforcements. There is much variation in their scope and content across countries and across disciplines in single countries.

Many funding agencies, institutions (e.g. universities) or scientific societies (e.g. the American Psychological Association, APA, or the German Psychological Society, DGPs) have already adopted their own policies for good scientific practice and handling of research data.

Recommendations on Safeguarding Good Scientific Practice

In Germany, the most prominent policy is the Recommendations on Safeguarding Good Scientific Practice which was proposed by the German Research Foundation (DFG, 2013). Many adaptations of these guidelines have been passed by universities or scientific organizations in Germany (e.g. adaption of the Leibniz Gemeinschaft). One of the core components of this guideline regarding research data is Recommendation 7: Safeguarding and Storing of Primary Data (p.21):

Primary data as the basis for publication shall be securely stored for ten years in a durable form in the institution of their origin.

Note the following implications of Recommendation 7:

  1. The storage duration does not depend on the data collection date but on the date of publications which are based on the data (even if your dataset was 20 years old, you would be required to store data if you published an article based on the data).
  2. Data shall be stored in a durable form (further information on long-term archiving can be retrieved from the knowledge base’s section on long-term archiving and data storage).
  3. Data has to be stored in the institution of their origin. This implicates that depositing data in an archive is not sufficient if there is no additional local copy.

However, in 2015, the DFG published an addendum on the handling of research data which introduces an appropriate nationwide infrastructure as an alternative to local storage.

In accordance with the rules of good scientific practice, research data should be archived in the researcher’s own institution or an appropriate nationwide infrastructure for at least 10 years.

Funding Agencies

Further information on funding agencies’ data management requirements can be retrieved from the knowledge base’s section on funding agency guidelines.

Institutional Data Policies

While institutional Research Data Policies are common in the UK (see  the DCC’s list of UK institutional data policies) or US (e.g. Briney, Goben and Zilinski, 2015), German institutions have been reluctant to take up their own research data policies. German universities only recently started to create research data guidelines of their own.

These guidelines vary across institutions. For example, while Heidelberg university’s guideline demands every research project to create a data management plan, HU Berlin’s guidelines do not even mention data management plans. However, HU Berlin’s guidelines establish legal security for researchers as they clearly state, that researchers are responsible for publishing their research data.

You can find a collection of German universities’ guidelines on

Journal Requirements

PLOS One is probably the most prominent example of psychological journals, that require researchers to make data underlying their analyses openly available. Other journals, that introduce similar requirements, exist, but are less well-known or have a narrower scope (e.g. Judgment and Decision Making).

Recently, journals are emerging, that exclusively publish results of pre-registered studies and require authors to share raw data upon publication (e.g. Comprehensive Results in Social Psychology).

Additionally, some APA journals recently introduced Data Transparency Policies (e.g. Journal of Applied Psychology) that aim to prevent duplicate or piecemeal publication of scientific findings.  These policies require researchers  to state, which variables of a data collection have already been included in published analyses.

Although no domain-specific overview of data policies in psychological journals exists, the following resources may be helpful:

Guidelines for Clinical Research

The ICH-Guidelines offer comprehensive policies, which are especially interesting for clinical psychology. Additionally, these guidelines can also be regarded as an example for a strongly formalized policy in a related domain.

The International Council for Harmonisation of Technical Requirements for Pharmaceuticals for Human Use (ICH) introduces policies regarding every aspect of clinical studies – from writing protocols and reporting clinical studies to coding of adverse events or performing statistical analyses. Hence, the ICH guidelines also touch important aspects of data management, especially in the range of documentation. In clinical studies it is common to  write down workflows in Standard Operating Procedures (SOPs, see the knowledge base’s section on transparent science) before the study starts.  You can refer to the website of the Technology, Methods, and Infrastructure for Networked Medical Research (short: TMF) to find sample SOPs.

Under certain conditions, further legal regulations, like the Arzneitmittelgesetz (German Drug Law) or Medizinproduktgesetz (Medical Devices Act), can apply to clinical psychological studies. Additionally,  a review of clinical studies through a medical Ethics Committee may be required (see the knowledge base’s section on Ethics Committees).

Further Resources