Gibbs, Graham R. (2004) Establishing an accessible data corpus based on evidence to the Victoria Climbié inquiry: Merging user needs and research dimensions. In: ESDS Qualidata Workshop: Online Access to Qualitative Data: Opportunities and Challenges, 15th January 2004, Royal Statistical Society, London. (Unpublished)
Abstract

The Victoria Climbié Inquiry was a major review of the child protection system in England and Wales and will set the direction for significant changes in that system for future years. The project at Huddersfield will establish a data corpus of the cross-examination evidence that was presented to the Inquiry and make it available on the internet.

The research involves three specific stages:
• identifying the themes and topics in the cross-examination that are required by or will be of use to a range of professional and educational users;
• cataloguing and coding the data and establishing a system of data management and retrieval;
• establishing a data corpus available for future research both within and outside the University.

The data corpus will use XML based standards for archiving qualitative data and the project is using Atlas.ti for data analysis and management as this offers the best current support for XML output. Although the data corpus has much in common with qualitative data archives from research projects, one distinctive feature is that the corpus will incorporate thematic elements and analytic features suggested by an investigation of user needs. Whereas for most archived qualitative data there remains the question of whether users (those undertaking secondary analysis) require details of the original analysis, in this project the thematic analysis will directly reflect user requests. This raises more broadly the question of how much of the range of ideas captured in the analysis of qualitative data sets (and usually recorded in the software used for analysis) needs to be incorporated into archiving standards. Several CAQDAS programs are now using XML or seem likely to. In the long run this will aid the transfer of data and coding from one program to another, but there are still some distinctive features that do not translate but that may be useful for an archiving standard (e.g. the ability in Atlas.ti to hyperlink quotations, or the ability in NVivo to create models (charts) that include multiple kinds of links to the same object). So if the archive is to keep some aspects of theoretical analysis as well as basic coding then there are still some difficult issues of storage format to be settled

Library
Statistics
Add to AnyAdd to TwitterAdd to FacebookAdd to LinkedinAdd to PinterestAdd to Email