Phenomenography is a research approach that focuses on phenomena and peoples’ understanding of phenomena. The principle areas of interest are identifying peoples’ conceptions of social and natural phenomena and documenting the ‘distinctly different ways’ in which people interpret the phenomena. It was developed in the 1970s by Ference Marton (1981) and has been used extensively in science education research (Prosser 1993 and Walsh et al 1993).
In this study, a phenomenographic approach to peoples’ understanding of scientific phenomena (quantum mechanics) has influenced part of the analysis phase within the Colaizzi (1978) framework. A phenomenographic approach is used to pinpoint and map variations in students’ understanding of quantum phenomena during the analysis phases. (Refer to Appendix 10 for a literature review on Phenomenography)
To understand student learning in quantum mechanics it is necessary to question practioners, researchers, teachers and learners of quantum mechanics. Interviews were selected as a principal data collection tool for both stages of this study. Interviews would provide opportunity for asking questions on a wide range of subjects, the possibility of probing issues more completely and exploring unknown issues as they arose.
Initially, estimates suggested that up to 100 subjects would be interviewed for this study. A task of this magnitude required a consistent interview protocol for the following purposes:
An in-depth interviewing technique was selected for this study, as it was best suited to the grounded theory and phenomenological approaches used in the two stages. Taylor and Bogdan (1984) give the following description of an in-depth interview:
“repeated face-to-face encounters between the researcher and informants directed toward understanding informants’ perspectives on their lives, experiences or situations expressed in their own words.”
The grounded theory stage of the study required in-depth data covering a range of ideas and examples in quantum mechanics, some of which were initially unknown. Subjects were to be engaged in a conversation and encouraged to articulate their understanding and models of quantum mechanics in their own words. Flexibility was needed to be able to diverge from an interview plan and explore specific issues when they arose. Issues that emerged in early interviews could be addressed in later interviews, allowing the flow of conversation to direct the research process. To fulfil these requirements the preliminary interviews were unstructured, or recursive, in nature; and progressively focussed toward a semi-structured format in preparation for the phenomenological stage.
The phenomenological stage of the study required in-depth data that was more focussed in its scope. The data from the preliminary interviews and other sources was analysed and used to produce an interview guide for the second stage. These interviews had many common elements, and dealt with specific issues brought to light by the grounded theory stage. The interviewer still required flexibility in the ordering and the depth of probing of individual questions but essentially the questions asked of each subject were the same. Interviews in the phenomenological stage were focussed, or semi-structured, in nature.
The interview questions were predominantly open-ended in format, and the subjects were also asked to draw or sketch as a part of their responses. The style of questioning was dominated by descriptive, structure, opinion and probing questions which are briefly described below (Table 3-2).
Table 3-2 : Description of Interview Question Types
To maximise internal validity the interviewer must constantly check perception and understanding (The interviewer’s and the subject’s) against possible sources of confusion. Probing, cross checking and recursive interviewing are techniques used to minimise this problem. For example, a common validity error can be as simple as asking the wrong question. Approaching the interview issue using multiple contexts, representations or examples can help as well as using other non-interview methods to get similar information (e.g. examination scripts, concept maps, drawings).
Reliability is increased by repetition. However in-depth, semi-structured interviews do not lend themselves (by their nature) to large sample sizes or exact duplication of interview conditions. Every effort should be made to improve reliability by careful planning of the interview, meticulous documentation of the interview conditions and process, accurate recording of interview data (e.g. note taking, tape recording etc) and wherever possible duplicating the interview conditions for each subject. This will also improve external validity.
A format was developed for conducting the interviews to ensure a consistent and pleasant experience for both subject and interviewer. A safe, non-threatening location for the interviews was selected and then fitted with suitable furniture, lighting and audio recording equipment. Subjects were invited to participate in an interview at a time convenient to them. Information about the interview was provided to subjects in advance. Before each interview the subject gave informed consent. The interviewer commenced each interview with questions to relax the subject and form a rapport. The interview was brought to a close with a series of reflective questions. The subject was thanked for their participation and provided with follow-up information on their interview.
All interviews were recorded on audio magnetic tape and later transcribed. Audio tape recording was chosen over field notes as a method for recording interview data for the following reasons:
It must be acknowledged that an audio tape record of the interviews also had disadvantages:
Directly following an interview the audio tape record and the interviewer’s memories and impressions had to be processed into a form that would allow complete and detailed data analysis at a later date. This form is called the transcript document and follows a suggested outline given by Ridge (1995).
The audio tape was transcribed and stored as a word processed file. A cover page was prepared containing details of the interview such as date, time, duration, location, interviewer, subject (coded to conceal identity) and key features that emerged from the interview. The cover page provided brief information that helps the interviewer recapture the events and feeling of the interview.
Following the cover page is the complete typed interview transcript, clearly indicating statements made by the interviewer and subject. It is situated in the middle of the page with a column on either side. On the left hand side is a personal log and on the right hand side an analytical log (refer to Figure 3-1).
Figure 3-1 : Example page from preliminary interview transcript showing layout
The personal log is an annotated chronological record of the interviewer’s reflections on the issues that arose during the interview/study. It is an instrument for recording reflective notes on the questions and responses, and it highlights areas for improvement and development in the interview process.
The analytical log includes reflective notes on the questions and responses, records emerging ideas and is a platform for discovering and organising ideas. Topics that are discussed or emerge are identified and connections made to the broader research questions. Trends or points of convergence usually emerge as the interviews proceed.
A detailed and accurate transcript document is produced for each interview and this allows analysis using different coding schemes to be overlaid as necessary. More details on the transcript document and its analysis appear in later chapters.