Research data management in economics journals: Data policies and data description as prerequisites of reproducible research

Replication of research results is eminent for empirical science. But in disciplines like economics, replication is a vision rather than a reality. One reason for this is that research data are not available due to the lack of mandatory data policies and archives. Even if data is available, descriptions with sufficient metadata are often missing. Also the e-infrastructure for providing datasets and other materials is still underdeveloped and offers no features. Our talk focuses on academic journals in economics. We present some results of a study of more than 140 journals regarding their research data management and suggest good practices for data availability policies. Subsequently we propose concepts for improvements regarding the journals' e-infrastructures. In particular we are addressing the problem of metadata creation. Often, the creation of metadata is not accepted by researchers because it is too time consuming. On the other side it must be comprehensive enough for reproducibility purposes. Referring to this contradiction, we define different levels of metadata schemata dependent on the different purposes they should serve - from ensuring the citation of research data to the requirements for replications of data and results.