Data Rescue : defining a comprehensive workflow that includes the roles and responsibilities of the research library

Abstract

This study, comprising a case study at a selected South African research institute, focused on the creation of a workflow model for data rescue indicating the roles and responsibilities of the research library. Additional outcomes of the study include a series of recommendations addressing the troublesome findings that revealed data at risk to be a prevalent reality at the selected institute, showing the presence of a multitude of factors putting data at risk, disclosing the profusion of data rescue obstacles faced by researchers, and uncovering that data rescue at the institute is rarely implemented. The study consists of four main parts: (i) a literature review, (ii) content analysis of literature resulting in the creation of a data rescue workflow model, (iii) empirical data collection methods , and (iv) the adaptation and revision of the initial data rescue model to present a recommended version of the model. A literature review was conducted and addressed data at risk and data rescue terminology, factors putting data at risk, the nature, diversity and prevalence of data rescue projects, and the rationale for data rescue. The second part of the study entailed the application of content analysis to selected documented data rescue workflows, guidelines and models. Findings of the analysis led to the identification of crucial components of data rescue and brought about the creation of an initial Data Rescue Workflow Model. As a first draft of the model, it was crucial that the model be reviewed by institutional research experts during the next main stage of the study. The section containing the study methodology culminates in the implementation of four different empirical data collection methods. Data collected via a web-based questionnaire distributed to a sample of research group leaders (RGLs), one-on-one virtual interviews with a sample of the aforementioned RGLs, feedback supplied by RGLs after reviewing the initial Data Rescue Workflow Model, and a focus group session held with institutional research library experts resulted in findings producing insight into the institute’s data at risk and the state of data rescue. Feedback supplied by RGLs after examining the initial Data Rescue Workflow Model produced a list of concerns linked to the model and contained suggestions for changes to the model. RGL feedback was at times unrelated to the model or to data and necessitated the implementation of a mini focus group session involving institutional research library experts. The mini focus group session comprised discussions around requirements for a data rescue workflow model. The consolidation of RGL feedback and feedback supplied by research library experts enabled the creation of a recommended Data Rescue Workflow Model, with the model also indicating the various roles and responsibilities of the research library. The contribution of this research lies primarily in the increase in theoretical knowledge regarding data at risk and data rescue, and culminates in the presentation of a recommended Data Rescue Workflow Model. The model not only portrays crucial data rescue activities and outputs, but also indicates the roles and responsibilities of a sector that can enhance and influence the prevalence and execution of data rescue projects. In addition, participation in data rescue and an understanding of the activities and steps portrayed via the model can contribute towards an increase in the skills base of the library and information services sector and enhance collaboration projects with relevant research sectors. It is also anticipated that the study recommendations and exposure to the model may influence the viewing and handling of data by researchers and accompanying research procedures.