The preliminary research project plan can be divided into two distinct stages. The first is the development of the set of interview questions and the second involves conducting and analysing the resulting interviews. Development of the interview questions required an open approach to generate a list of options, and which allowed exploration of various possibilities before settling on the final set. The interview process required an approach which was flexible and responsive, and would allow the researcher to become immersed in the analysis process throughout the data collection.
To meet these research objectives two approaches were selected and adapted for this study. For the development of the interview questions a Grounded Theory approach was chosen and for the interviews a Phenomenological approach was adopted. The utilisation of interviews formed a major common component across both activities, and therefore a robust interview protocol was necessary to ensure that interview data was recorded and presented appropriately. The research was conducted in a social context and it was recognised that the complexity of human behaviour requires the triangulation of both data and methods to increase validity and reliability. A description of these methods and their adaptation to this study are described in the following sections.
3..1Grounded Theory Approach – Stage 1
Grounded theory is a qualitative approach to data collection and data analysis. The approach was developed by Barney Glaser and Anselm Strauss in the 1960s and has its origins in sociology. By the late 1980s Glaser and Strauss had moved in differing theoretical directions (Melia 1996, Stern 1994). For this study the perspective described by Strauss and Corbin (Strauss and Corbin 1998) was adopted.
The aim of grounded theory is the generation of theory from the gathered data, although for the purpose of this research project the grounded theory approach was utilised to identify a list of key elements and generate a set of interview questions. These interview questions formed the starting platform which launched the phenomenological stage of the study.
Grounded theory was selected for the first stage of the study for the following reasons: results generated are based directly on the data; the research approach starts with an area of interest, then data is collected and relevant ideas are allowed to develop; the approach encourages the researcher to collect data from multiple sources and guides the researcher to examine the data from all sides rather than staying fixated on the obvious; the researcher compares each section of the data with every other part throughout the study looking for differences, similarities and connections; and the approach is considered especially useful in situations where little is known about the subject under investigation.
Developing Theoretical Sensitivity
The researcher’s theoretical sensitivity provides a platform for multifaceted examination of the collected data rather than becoming fixated on the obvious. In this study the researcher informed and developed his theoretical sensitivity throughout the research project by reviewing related literature, seeking discussions with a wide range of practitioners in both teaching and research settings and through the processes associated with data management and analysis.
In grounded theory the processes of data collection and analysis are intertwined; as the data is collected from various sources it is analysed and compared with all other data to identify trends, differences, similarities and connections.
Grounded theory does not begin with a hypothesis. Instead, after the initial data collection and analysis, relationships are identified and provisional hypotheses can be formulated, which are then checked against further data and modified. Researchers are encouraged to utilise the widest selection of data sources possible which, for this study, included observations, interviews, examination scripts, diaries, literature and curriculum related documentation.
For this study this recursive process allowed the researcher to formulate, test, modify, review, discard and develop many provisional hypotheses within the context of the collected data. These processes increased my theoretical sensitivity and assisted in the final formulation of the interview questions.
The sampling process employed in grounded theory is driven by theoretical, rather than purely statistical, considerations. This provides the researcher with a responsive research tool and allows for additional leads to be adequately investigated. In this study, theoretical sampling allowed me to target specific types of students, which in turn verified several aspects under investigation by allowing categories to reach an obvious point of saturation. These may have been inconclusive or overlooked if a predetermined sampling regime had been employed.
As previously discussed data analysis in grounded theory is an iterative, continual and integrated process performed throughout a research project. The initial analysis consists of coding and categorising, and usually involves a line by line analysis of transcript style documentation. The key feature of the codes and categories generated by this analysis is that they are based directly on the data.
Coding in grounded theory comprises three steps, elaborated below. The first two steps form a cyclic system and the researcher alternates between the two processes of ^ processing for the incoming data. The third step of Selective Coding attempts to identify a core category which links all other categories together.