Karstens S.1, Hormesch T.2, Blaesius K.H.3
1Trier University of Applied Sciences, Therapeutic Sciences, Trier, Germany, 2Trier University of Applied Sciences, Student Computer Science, Trier, Germany, 3Trier University of Applied Sciences, Computer Science, Trier, Germany
Background: Academic journals are increasingly demanding the deposition of data underlying the findings of research manuscripts in a public repository. This advance simplifies reanalyses including meta-analyses, and the development of new research methods. Moreover, easily accessible data can be used for training programmes in statistics. There is a range of repositories available, but searching for data is currently cumbersome, due to the spread deposition.
Purpose: Development of a meta-search engine for research data repositories.
Methods: A search engine for deposited research data (SeaRes), was programmed using the Common Lisp language. The websites of common databases indexing research articles were analysed before developing the graphical user interface (GUI) enabling the combination of search terms using Boolean logic. Filename extensions were added as searchable items due to their specific relevance with regard to data-repositories. A general internet search was performed non-systematically to identify relevant repositories, and an All Fields test run on two repositories (Dryad Digital Repository and Figshare) searching for the term physiotherapy was performed.
Results: A prototype of the search-engine accessing Dryad Digital Repository and Figshare is working. The test run resulted in 110 hits for physiotherapy (October 27 2016; 9 Dryad, 101 Figshare). As a result of the search for additional repositories Harvard Dataverse Network and Zenodo will be incorporated in the next step.
Conclusion(s): Openly available data is currently limited, but it is expected that the amount of data repositories and datasets will rise enormously. After it is finalised, the SeaRes-tool will make it easier to find data of interest without visiting numerous websites. Because of the rapid evolvement of academia and scientific databases in particular, future technical adaptations will be necessary. Moreover, the user friendliness of the tool will be further improved, and may be made available as an open source product.
Implications: In the future, with the tool at hand it will be possible to further develop meta-research methods for physiotherapy research. The tool may also prove to be a valuable aid for the education of physiotherapists. Using SeaRes, it is possible to easily identify datasets which can be used in training-programmes in statistics. The establishment of an international user community is intended, to foster further developments.
Funding acknowledgements: The work was unfunded.
Topic: Research methodology & knowledge translation
Ethics approval: Ethics approval was not required.
All authors, affiliations and abstracts have been published as submitted.
