Meeting the Long-Term Needs of Scientific Progress Through Data-Collection Standards

• Activity Number: 234
• Type: Topic Contributed
• Date/Time: Monday, August 5, 2013 : 2:00 PM to 3:50 PM
• Sponsor: Business and Economic Statistics Section
• Author(s): Richard Welpton*+ and Matthew Woollard and Melanie Wright
• Companies: UK Data Archive and UK Data Archive and UK Data Archive

Abstract

Access to potentially identifiable data are typically restricted: fortunately, secure access mechanisms which protect the identity of data subjects are now well established. However, standards for producing properly curated data and metadata are far from satisfactory applied. Long-term consideration of this issue is essential, not only for replicability of results, but also for users of data with long-term research interests. Data custodians with responsibilities for identifiable data are more willing to concentrate on the potential effects of the disclosure of such data without considering the long term consequences of cost implications, lost research and even lost data. This paper will argue 1) that data custodians should particularly consider long-term curation of data collections which are potentially identifiable as a key responsibility, 2) that data access providers can generally follow well-established principles with restricted-access data but with additional requirements for information security and 3) that secure access methods improve the levels of replicability which are possible within data which is not identifiable.