Applied hermeneutics and qualitative safety data: the CIRAS project

This article describes the new qualitative methodology developed for use in CIRAS (Confidential Incident Reporting and Analysis System), the confidential database set up for the UK railways by the University of Strathclyde. CIRAS is a project in which qualitative safety data are disidentified and then stored and analysed in a central database. Due to the confidential nature of the data provided, conventional (positivist) methods of checking their accuracy are not applicable; therefore a new methodology was developed - the Applied Hermeneutic Methodology (AHM). Based on Paul Ricoeur's 'hermeneutic arc', this methodology uses appropriate computer software to provide a method of analysis that can be shown to be reliable (in the sense that consensus in interpretations between different interpreters can be demonstrated). Moreover, given that the classifiers of the textual elements can be represented in numeric form, AHM crosses the 'qualitative-quantitative divide'. It is suggested that this methodology is more rigorous and philosophically coherent than existing methodologies and that it has implications for all areas of the social sciences where qualitative texts are analysed.