Potential Impacts

This project will form a case study of how technologies (such as computational linguistics) can be applied to the practices of social science data archiving, presentation and sharing. One indicator of an impact already achieved is the bringing together of scientists across sites to conceptualise and write this application. Enabling e-qualitative data is a prerequisite for e-science. Persuasion and encouragement through promotion and outreach is one of the better ways of getting the evolving e-paradigm to become accepted. This project will help contribute to this mission, and will build on the networking role of the UKDA, as an ESRC Research Resource Board provider, in helping reach out to social researchers, data creators (both converts and sceptics of archiving and sharing), and data users. Edinburgh too, are well placed to convince computational linguists that social science data is worth attending to in addition to their current work with web pages, publications and news streams.

More specifically, the project can be expected to have longer term impact across the following areas as set out in the bid:

1. developing an area of theory or methodology. The project will develop a standard, methods and tools for preparing, annotating, documenting, linking and presenting (eg web-publishing) qualitative data. NLP tools will be applied to help automate some of the lengthy manual processes;
2. collecting a new body of information or data; and contributing to knowledge or understanding. The project makes use of test data from research-active sources that are currently being re-used. The reports and publications produced will provide three sets of innovative evidence and practical guidance in the areas of: applied investigation of social science methodology with respect to secondary analysis of qualitative data; computational linguistics as applied to qualitative research data; and e-science considerations for grid-enabling qualitative data. Working exemplars that show how social scientists can conceptually and practically link and merge data is much needed, and indeed we believe this to be a fundamental prerequisite for key e-social science applications;
3. developing research methods or techniques. The project, in its investigation of the role of contextuality and representation of qualitative data will produce guidance on best practice in capturing context through documenting and annotating data for researchers and for data producers;
4. contributing to policy or practice in certain areas. The tools, guidance and outreach, and capacity building aspects of the project will contribute to practice in the areas of data sharing for data distributors or those working to facilitate applications for qualitative data analysis, and are as relevant to non-academic organisations and commerce as they are to the academic domain. The collaboration with qualitative software companies in a plea for them to subscribe to non-proprietary and shareable data standards could be seen, ideally, as having huge potential benefits for researchers. The project's contribution to the outreach work of the QUADS Coordination will also function alongside ESDS's efforts to promote data use at a variety of levels;
5. providing a product such as computer software, patents or research facilities. In addition to enhancing our knowledge of how and when data can be best exposed for sharing, the project aims to deliver a suite of tools, which while they may not be shelf-ready release versions, will provide a starting point for data disseminators and hopefully, be of interest to the e-science community. We believe that in order to be of longer-term benefits to the community, it is critical that these toolsets are well tested across the potential range of qualitative research specialities and applications, and are accompanied by step-by-step guides and tutorials on how and how not to use them;
6. publishing books and articles and project website as dissemination. The project outputs will go beyond diffusing into the formal literature. Findings and reports in the form of reports and guidance (as documents or web pages) will be integrated into the UKDA/ESDS web site. As such, the longer-term archival orientation of ESDS means that this information will remain visible in the future. We would anticipate the e-social science centre and its future incarnations will provide access to the tools and reports arising out of this project;
7. longer-term capacity building. The projects grander aim is to inject a new technological spark into the thinking and practice of how we can share qualitative data. The take up of new approaches needs to be accepted and incorporated into mainstream methodological thinking and practice for qualitative social research. Through the UKDA involvement in the educational and capacity building roles of ESDS and the NCeSS, this project will help to promote and inspire the uptake of such methods and tools as part of the centres' own strategies. The creation of formalised links between the UKDA (through ESDS Qualidata) and the NCRM will be expected to firmly place this project's developments into the longer-term ESRC methodological arena;
8. new avenues of collaboration. This project brings together research expertise and applications from social science research and methodology with computational informatics/linguistics as applied to qualitative data archiving and sharing. Evidence of successful bridges between these two disciplines are sparse, and need to be developed, particularly if we are to fully realise the potential that e-science can bring to social science data provision and analysis. We envisage that through the extension of, and application of, real social science data to existing computational linguistic methodologies and tools, a practical contribution to interdisciplinary collaborative practice and innovation will be made.