Method is about selecting individual techniques to collect information (Blaxter, Hughes and Tight 2004: p.59; Easterby-Smith, Thorpe and Lowe 2002: p.31). From a qualitative perspective, the selection of method(s) must be appropriate for capturing deep and detailed data (Patton 2002: p.14) and is thus suited to open-ended interviews, direct observation or methods derived from the use of written documents (Labuschagne 2003). As this research adopts a phenomenological stance, interviews and observation are the most suitable methods (Sanders 1982).
Techniques commonly associated with quantitative data, such as statistical analysis which is used to make generalisations about the characteristics derived from a sample population (Cooper and Emory 1995: p.116), measurement and scaling, where mathematical properties are typically used to attach values to findings and data reduction which reduces numerical or alphabetical information to patterns, are all discarded. Moreover, Sander's (1982) assertion regarding the choice of techniques also implies that qualitative methods (apart from interview and observation) that comprise the social research design framework have limited application for phenomenological research, and for this reason, are also dismissed.
Techniques such as conversation analysis (that involves analysing complete transcripts of conversations (Babbie 2007: p.383)), narrative (the analysis of personal stories (Fraser 2004)), case study (the method of organising data by case for in-depth study and comparison (Patton 2002: p.447; Hartley in Cassell and Symon 2004: p.323)) each have some bearing on this research. Therefore, relevant aspects of these stances will be considered in the research design. As stated, observation is relevant for phenomenological research (Sanders 1982) and encapsulates a variety of systematic and selective (Kumar 2011: p.140) activities including observing, documenting and analysing phenomena (Marshall and Rossman 2011: p.139; Blaxter, Hughes and Tight 2004: p.178). It is, however, considered less appropriate for this research undertaking because the key concepts under study e.g. VfM, partnership management, risk management and performance management cannot always be directly observed or experienced (Patton 2002: p.22-23) due to their abstract nature.
The remainder of this section discusses the sampling technique selected for this research. It also presents the main methods that will be used for collecting data, consisting of a two-phase approach. In the first phase, secondary data will be used to identify issues that will inform the development of a conceptual PPP management model for partnership, risk and performance management. This model will be developed from the literature review and case study findings (Chapters 3, 4 and 5) through an integrating process (Chapter 6). Primary data will then be collected to validate the model, through feedback obtained from selected industry representatives (Chapter 8). The second phase of the research will use this feedback to create a second iteration of the model that will be subjected to focus group validation. Primary feedback from the focus group will then inform a third and final iteration of the model (Chapter 9).