Quality Assurance

The ADA expects a data deposit to be of a certain quality when it is submitted (see 2. Deposit Preparation for instructions). Nevertheless, all data deposits are assessed for quality by an ADA archivist. This assessment includes both the content and the form. On the content side, the ADA archivist will scrutinise the data for direct and indirect identifiers. On the form side, the ADA archivist will check that all variables have un-ambiguous clear labels for all variables and runs spell checks, as well as basic consistency and completeness checks. The ADA archivist will propose changes to the depositor. If the depositor does not agree to changes the ADA archivist deems absolutely necessary, a deposit can be rejected. The ADA archivist will implement the agreed upon changes and furthermore generate versions of the data for SPSS, SAS, STATA and CSV for maximal usability.

Accompanying documents are uploaded to ensure comprehension of the study and the data sets. If no such documentation accompanies a study the archivist will liaise with the depositor to ensure all necessary value labels and codes are defined, that metadata fields can be completed in the DDI and that the data is understandable to other researchers.

It is not a requirement that a study meets all fields of the DDI but it is imperative for it to meet a minimum requirement. This minimum requirement is not published in the public domain but is referred to as DDI lite within the archive.

Related publication and project websites can be linked in Dataverse to give context to a dataset.

Apart from that, the ADA uses APAIS (http://vocabularyserver.com/apais/) vocabulary for keywords and classification in the Dataverse catalogue to improve findability.