Conceptual and Methodological Issues
Chapter 1 described the intention of the Thames Valley study to unite different theoretical approaches to satisfy a range of aims. An eclectic approach to methods was deemed to be necessary in order to achieve theoretical triangulation. Hence, I chose not to restrict myself to using either positivist or interpretative methods, but to adopt a pragmatic approach to data collection. It was decided to use any appropriate method in as accurate a way as possible: to unite holistic analysis, which is characteristic of the positivists, and atomistic analysis, characteristic of the interpretive sociologies. This pragmatic approach resulted in the use of triangulation techniques; both 'methodological triangulation' and 'data triangulation'.
There are two forms of 'methodological triangulation'; 'within- method' and 'cross-method'. The former involves the use of differing strategies within a broad research method. In this study the use of semi-structured interviews with officers which generated quantitative data and qualitative descriptions fits the 'within-method' type. 'Cross-method' triangulation refers to the procedure of using different types of methods to study the same phenomenon. Interviews with police officers and victims, observation of officers on duty, and the examination of official records were all used to understand (amongst other things) the police response to incidents of domestic violence.
The use of different types or sources of data within one study is known as 'data triangulation' (see Denzin 1970). This study generated data sets on police officers, prosecutors, and victims in order to consider what the role of the criminal justice system is, and should be, in relation to victims of domestic violence. Triangulation not only allows for examination of the various facets of a given