Where Does All the Data Go: Quantifying the Final Disposition of Research Data

Abstract

This paper describes the results of a large survey designed to quantify the final disposition of research data and to determine the extent that the key factors of repository technologies and data management resources influence the decisions that researchers make about their data. A total of 724 National Science Foundation awardees completed this survey. Removable media devices such as CDs, DVDs, and removable hard drives are the most common storage technologies. Less than 30% of researchers use community or reference repositories. More than half of the researchers get storage through their institutions. Institutional support for storage is significantly higher than for data management. The key finding of this research is that the availability of resources is the most significant indicator of whether the final deposition of data is in a preservation‐worthy manner. Resources, provided by the research institution or via grant funding, include storage technologies, repository systems, and data management personnel. When high quality data storage is offered, researchers will use it. When professional data managers are available, data is more likely to be stored in more sustainable technologies.

Before the Repository

“…In the literature on preservation, the rapid rate at which removable media becomes obsolete are discussed [31]; however, the loss of the physical media is not treated as a threat. As large percentage of researchers copy their data to removable media such as CDs, DVDs, and hard drives at the end of a project [19], removable media management needs to be included in the discussion of threats to preservation.…”

“…The primary threat mitigation strategy is good data management [14,19,25,36]. Data management is a set of tasks and activities to plan for system backups, contingencies, process resumption, hardware and network redundancy, automatic failover, and site mirroring -all efforts to avert problems.…”

“…However, only a small percentage of research data is submitted to repositories for long-term preservation [19]. Individual scientists are responsible to manage their own data [24,25].…”

“…The probability of long-term data management for research collections is low when the ongoing responsibility lies with an individual researcher or graduate student [19,28].…”

“…Many lifecycles make an assumption that data is eventually stored in a preservation infrastructure; that is, data is ingested into a digital repository. However, only a small amount of research data is actually managed in a preservation environment; most research data is maintained by the researcher [19].…”