|[ Abstract for this presentation ] [ Proceedings Contents ] [ Schedule ] [ Abstracts ]
Devising a phenomenological study within the qualitative paradigmJohn Bednall
The University of Notre Dame Australia
Sixteen Heads of Anglican and Uniting Church independent schools were asked to describe their experiences of providing leadership to religiosity in their schools. It was found that there was a common but strong ironic tension characterising how each of the affiliated Churches engaged with the religious culture of their schools . The authenticity of this research was established through a carefully designed model of phenomenology. A valid distinction between the notions of epoche and bracketing was articulated and then a process for the clear operationalisation of both was designed. The professional literature seems reluctant to be as specific in describing either concept within qualitative research methodology. This presentation will describe how the researcher activated both and how the intuitive connection of the researcher's own life experiences to the data provided by the sample Heads could be authenticated thus demonstrating the uniqueness of phenomenology as a social research method.
The primary data collection device was a semi-structured interview with each Head with questions fitting either the main, probe or follow-up categories described by Rubin and Rubin (1995). The main and probe questions were asked at face-to-face interviews with follow-up questions or member-checking devices (Punch 2004) being conducted by email correspondence. A careful interview protocol was designed based largely on the work of Wengraf (2003).
In contemplating a methodology, it was clear from a pre-research review of the scholarly literature and a preliminary inquiry based upon conversations with a range of Heads of schools of different religious affiliations, that a qualitative framework was apposite. The possibility of a phenomenological study emerged because of its perceived capacity to process authentically the subjective and the value-laden from a small, purposeful, non-representative sample group.
There are two conceptual propositions supporting phenomenological method within qualitative research design. The first is that the reality of a set of human experiences will be uncovered through the detailed yet subjective descriptions provided by the people being studied (Creswell 1994) and second, that 'establishing the truth of things' (Moustakas 1994, p.57) begins with the researcher's perception. The researcher is able to rely upon 'intuition, imagination and universal structures to obtain a picture' (Creswell 1998, p.52) of the experiences under study. I was therefore attracted by the possibility that my own experiences as the Head of both an Anglican and a Uniting Church school could be used authentically to enhance the significance of the proposed study.
Additionally the research report was written in the first narrative person. If the researcher and the research respondents as living human beings are at the core of phenomenology, it seems dysfunctional and illogical for the research findings to be de-personalised by the use of the third narrative person. As the narrator in this study, I was legitimately also an active agent in establishing with the respondents the dependable research findings. Conscious that it not only confronted the formal and impersonal tone of positivism as well as the objective, interpretative styles of much social inquiry, including those of the qualitative tradition, I decided a phenomenological research report reflected a more authentic outcome if written in the first narrative person.
When I was Head of two independent church schools, it became very clear to me that my leadership of the religious domain of two complex Church schools was operating within uncertain and ill-defined theoretical and empirical guidelines. I knew that the governing Council of both my schools had made certain, but not extensive, inquiries into my religious views before my appointment as Head was announced. I knew both the schools declared their denominational affiliations and made public statements in their promotional materials that they were learning communities where 'Christian values' were extolled. However, I was never entirely confident I understood the affiliated Church's expectations of me or to what extent I had formal accountability to the presiding senior figure of the Church. I was also unaware of any clear guidelines from either of my governing Councils for my leadership of religiosity. Even the descriptor 'Church school' began to assume for me of meanings which did not helpfully guide professional and public discourse associated with such issues as religious and values education or even the constitutional validity of government funding. The relationship between my schools and their Churches seemed to need a mutually supportive re-definition, especially if any aspiration to be 'denominationally distinctive' (McLaughlin 1996, p.320) was to be realistic.
Hearing the voices of a group of contemporary Heads of Church schools as they told their personal stories or related the 'epiphanic moments' (Angelides 2001, p.430) of their leadership experiences, and then interpreting their accounts through a phenomenological integration (Gearing 2004) in which my own experiences were germane, provided sharp insight i nto the uniqueness of the experience of leading religiosity in an Australian independent Church school of the Anglican and Uniting Church tradition.
To understand both concepts the structural relationship between a researcher and the research respondents, merits explanation. This relationship is in place at the very beginnings of a phenomenological inquiry. The challenge for a researcher is to allow the voices of subjectivity to emerge authentically in coming to an understanding of what essentially the research respondents mean in their personal accounts expressed in the data collection devices. This placed upon me as the researcher in this study, the obligation to separate any past knowledge or experience I might have had in the leadership of Anglican and Uniting Church schools and then to use that experience by connecting it interpretatively to the meanings of the respondents. A connected relationship was only made possible by the concepts of epoche and bracketing.
The literature has generally treated bracketing and epoche as interchangeable or synonymous (Gearing 2004; Beech 1999; Ray 1990; Spiegeberg 1973). Gearing (2004) however, declared there are some underlying philosophical differences between the two terms, but they can be described or defined interchangeably 'to reflect the similarity of their core essences' (p.1430). But the definition and activation of both epoche and bracketing often lacks uniformity of standards particularly when critical engagement occurs with the issue of researcher subjectivity in data interpretation. Gearing (2004) commented:
Many ensuing phenomenologists who accepted [bracketing] as essential to the tradition... interpreted or described this concept individually to fit with their respective phenomenological writings (p.1431).The literature is not forthcoming in describing the activation of bracketing or epoche in phenomenological inquiry or providing exemplars of operations which demonstrate the concepts at work in a research project. Furthermore, there is some scepticism expressed by Ashworth and Lucas (2000), Colaizi (1978) and Porter (1993) that it is impossible to attain the degree of objectivity required for authentic epoche/bracketing if a researcher has had experience of the phenomenon under attention. Crotty (1996) Groenewald (2004) Schultz (1994) also raise the additional question of how researchers deal with the differing intensity of experiences over time and within changing contexts. Even individual respondents in a study may provoke varying emotional impacts upon the memory of a researcher of the same event. Myerhoff and Ruby (1992 in Ahern 1999) prefer the term reflexivity whereby the researcher seeks to understand the impact of personal experiences on data interpretation rather than engaging in futile attempts to eliminate it.
Gearing (2004) is not alone in identifying practical distinctions between epoche and bracketing. Variations of function and purpose are also argued by such researchers as Ahern (1999), Denzin and Lincoln (1989), Groenewald (2004) and Patton (1990). The distinctions emerge from how a researcher engages with data at the pre-empirical, collection stage and how that engagement shifts at the post-empirical interpretation stage. Patton (1990) described epoche separately from bracketing as 'an-ongoing analytic process' (p.408), which implies it should be dynamically integrated into the sequential progress of the whole research method from the very beginning of the study. Acts of bracketing, on the other hand, would occur at those interpretative moments when a researcher holds each of the identified phenomena up for serious inspection. A researcher then allows those personal ideas and feelings held in epoche to synthesize with those observations as interpretative conclusions. This is the process described by Gearing (2004) as 'reintegration' (p.1434) which consists of '...the unbracketing and subsequent reinvestment of the bracketed data into the larger investigation' (p.1434). Epoche, accordingly, allows for empathy and connection, not elimination, replacement or substitution of perceived researcher bias. Bracketing advances that process by facilitating a recognition of the essence of meaning of the phenomenon under scrutiny.
My major challenge then in seeking an authentic phenomenological method, was the design of a particular mechanism which enabled me to be aware of my own potential for bias, how to suspend that bias at the commencement of data collection, and then use an explicit process to evaluate the significance of that bias in data interpretation. There needed to be a functional symmetry between myself and the research focus and then a structural relationship between myself and the data under examination.
Such symmetry was first established by considering the etymology of the word epoche. The word appears to be Greek, thence Medieval Latin, the form being epock meaning a check or a pause (Macquarie 1997). In modern English, the form is epoch meaning 'a particular period of time marked by distinctive events' (Macquarie 1997, p.716). Accordingly within qualitative research, epoche can reasonably be interpreted as highlighting a particular period when significant events occur in the experiences of a researcher, but any impact from the memory of which need to be put aside during data collection. It therefore contains connotations of continuity and sequence, as opposed to the single act or episode of bracketing which would occur at or immediately prior to data interpretation. The symmetry of the intellectual dynamics inherent in the application of epoche and bracketing can be illustrated figuratively:
This figure (above) attempts to illustrate how a balance is operating between a researcher's awareness of past experiences and the units of meaning identified in the data accounts of the research respondents. Each is kept apart from each other or held in suspension by epoche. The two states are taken through to reintegration when the items held in epoche are assessed for any synthesis with the flagged items data collected from the respondents. A mathematical allusion is helpful in understanding this process. A formula is devised by adding to the flagged items - but holding in brackets - the items of recall identified in epoche. This revealed how the bracketed data would stand in relation to the respondents' data in order to suggest conclusion. Gearing actually used the term 'unbracketing' (p.1434) to describe the event of "removing the brackets" which leads to a fusion between the two sets of information. The idea expressed in this fusion or "unbracketing" emerges as the interpretative statement. Epoche therefore is a habit of thinking which continues throughout the pre-empirical and post-empirical phases of the study. Bracketing is an event, the moment of an interpretative fusion and the emergence of the conclusion.
The Feelings Audit reflected the stark reality of what makes phenomenological inquiry authentic: with all of my personal dispositions and values, as the researcher, I was at the centre of the interpretative process. The Audit was taken to each research interview to assist me in some personal reflection. It was not shared with the respondents. Its major purpose was to facilitate epoche by persistently reminding me of items where consciously I had to set aside any value judgments of items of data contributed by the respondent Heads.
Phenomenological theory often alludes to the seeking out of units of meaning. However, the term has an implicit pre-suppositional tone to it, whereby an item of relevance to the research area appears to have 'meaning' assigned to it in terms of its potential significance to interpretative conclusions yet to come in data analysis and hence the events of "un-bracketing" (Gearing 2004, p 1434). In order to protect the integrity of epoche as the on-going process of avoiding any such pre-suppositions at all costs, the term flagged item is used later in this article in the description of a six phase data interpretation process. However, the definition of units of meaning provided by Ratner (2001) is consistent with the treatment afforded flagged items in this study:
One word may constitute a meaning unit. Several sentences may also constitute a unit. A meaning unit may contain a complex idea. It simply must be coherent and distinctive from other ideas. The meaning unit must preserve the psychological integrity of the idea being expressed. It must neither fragment the idea into meaningless, truncated segments nor confuse it with other ideas that express different themes (p.2).
The research journal became deeply entwined in data interpretation as well as functioning as an aide-de-memoir which Miles and Huberman (1984, p.69) called memoing, an important data management process in qualitative research. Some of these themes and ideas emerged as a result of my own background and experience in the research area and were noted as such. The journal entries were dated but were not shared with the respondent and as such maintained the distinction between a researcher's theory language and the respondent's interview language (Wengraf 2001). The content of the research journal was not used to establish a comparative framework across all the interviews. The phenomenological intention of the research was to hear the distinctive stories of each of the respondents. It was only at data interpretation that any shared or common themes in those descriptions became active components in my thinking.
I found keeping my personal views and experiences separate from data collection more challenging than I had anticipated. Simply utilising epoche as a habit of thinking appeared not to be enough. The mechanism required reinforcement through some practical measures. Accordingly any personal allusions to past experiences or possibly prejudicial comments I made during the interview conversation, which had potential to be pre-emptive, were not included in the interview transcript to protect the distinctive and personal voice of the Head's account. However a brief summary of any such allusions or observations was made in the research journal so that all subjective comment was held outside data collection until the event of data interpretation and a decision could be taken as to whether they augmented or reduced the significance of data. A cumulative record of quoted text from each Head's interview transcript was also kept and progressively graphed (see below). This enabled me to detect if any one Head was assuming unwarranted dominance in the interpretation of data, or indeed if I was not paying enough attention to any other Head.
Percentage of Transcript Text quoted in thesis as support data
Source: Audited Transcripts of Research Interviews, conducted June/July 2004
Such issues of balance in the use of interview data is alluded to by Eichelberger (1989) in his warning that phenomenological evidence is the product of commonality of experiences by the respondents collectively, rather than through the dominance of any one individual story. My study was not to report upon sixteen differing stories of church school leadership but to identify where there were common and shared experiences and to unite them as a single narrative about the nature of the experience. Although there were two examples in the interpretative sections of the research report where the example of one particular Head was used to exemplify an important topic of significance on behalf of the entire sample group, it is not considered that the desired balance in contributions by all the Heads was compromised as the graph (above) demonstrates.
Redundant units, that is those not assigned to a topic of significance, were not discarded from the data but transferred to a miscellaneous category in case their significance to the holistic impact of any one Head's account, or to the phenomenon, became apparent. Minority voices were thereby retained because they are 'important counterpoints to bring out regarding the phenomenon researched' (Hycner 1999, p.154).
Thirteen Topics of Significance were decided upon with an additional grouping of Miscellaneous being the flagged items which were not coded. I then satisfied myself that these objective categories could be reasonably confirmed as functioning responses to the main questions at interview as well as within the ambit of the research intention and the research questions. This was consistent with processes developed by Weston, Gandell, Beachamp, McAlpine, Wiseman and Beachamp (2001) in their phenomenological study into the characteristics of university teaching, whereby the starting point for the design of a qualitative data coding system was to establish the relationship between such a system and the primary research questions. The topics of significance were listed under the major contextual issues according to the following table.
|Major Contextual Issue||Topics of Significance|
|The Head and Leadership|
|The school praxis|
|Source: Transcripts of research interviews with sixteen sample group Heads|
The orientation standpoint involved my epistemological and ontological stances as the researcher. As a result of acknowledging my past experiences in independent Church school leadership and my awareness of their impact upon my perception of a personal reality, I was able to flag references within each of the written transcripts of the interview as being of potential consequence to the research objective. This was never seen to be a breach of epoche because all assumptions about causations, consequences and wider significance were suspended. Epoche disallows 'constructivist meanings' (Gearing 2004, p.1437) to be attached which have potential to pre-empt later interpretations. To discern an item to be of potential relevance to an openly stated research intention is not to infer any conclusions about meaning.
The second element of praxis involved the research analysis itself. Flagged units were extracted from the written transcripts and clustered together on the basis of their having common elements related to the Heads' experiences. Each group or cluster was given an objective coding title as a topic of significance. To have been included, each flagged item had to have a systematic relationship to the questions at the research interview. They had to have been indicated by the Heads as specific elements of thought which were compellingly represented in their shared and expressed experience of leadership of religiosity.
The third element involved the reintegration of the r esearch praxis with the research objective. It therefore included the final and essential phase of bracketing. The topics of significance were categorised within emergent contextual areas, which provided the headings and the content for each of the interpretative chapters of this thesis. The subsequent 'reinvestment of the bracketed data into the larger investigation' (Gearing 2004, p.1434) was achieved by dismantling the bracketed ideas and bringing the released data into reintegration with the Heads' accounts.
It was fully appreciated at the outset of this phenomenological exploration, that the research objectives would venture into a complex scenario of human values and attitudes, nuances of definition relating to the inherently emotional subject of religious witness, and a range of ideas and propositions that would be difficult to express as single and coherent themes. The research design, and in particular the typology for epoche and bracketing outlined in this article appear to have had dependable outcomes, indicated by the fact that I found it relatively straightforward to isolate ten major conclusions from the study all of which sat clearly within the boundaries of the initial research questions.
The conclusions express the realities of shared and common experience which could only have emerged from the non-rational and subjective descriptions contributed by the Heads, or as Loader (1997) had hoped, from their "personhood of ... and the emotions [they] experience while leading' religiosity in their schools". (p.3). All of the conclusions (10) sat clearly within the boundaries of the research questions and illustrated the personhood of the respondents. Key words in the expression of the conclusions spoke of human relationships, personal philosophies and chosen preferences, the need to seek understanding and support in the onerous challenges of leadership, the empathy felt by the Heads for the full extent of stakeholders in the school community, the inculcation of values, the reasonableness of expectations within the full praxis of religiosity of the staff.
Such is the power of phenomenology.
Ashworth, P. & Lucas U. ( 2000). Achieving empathy and engagement: a practical approach to the design, conduct and reporting of phenomenological research. Studies in Higher Education, Vol 25, Iss 3, pp 295-309, Abingdon, UK.
Beech, I. (1999). Bracketing in phenomenological research, quoted in Gearing, RE 2004, Bracketing in research: a typology, In Qualitative Health Research, Vol. 14, No. 10, December 2004, pp.1429-1452, Sage Publications, Newbury Park, CA.
Colaizzi, P. F. (1978). Pyschological research as the phenomenologist views it, in Ahern, K.J. Pearls, pith and provocation: ten tips for reflexive bracketing, p. 408, Qualitative Health Research, Vol, 9 No 3, May, 1999.
Creswell, J. (1994). Research design: qualitative and quantitative approaches. Sage Publications, Thousand Oaks, CA.
Creswell, J. (1998). Qualitative inquiry and research design: choosing among five traditions. Sage Publications, Thousand Oaks, CA.
Crotty, M. (1996). Phenomenology and nursing research. In P. Barkway (2001), Michael Crotty and nursing phenomenology: criticism or critique? Nursing Inquiry, Vol 8 Issue 3, pp. 191-200 [accessed 15 Feb 2005] http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&list_uids=11882218&dopt=Abstract
Denzin, N.& Lincoln, Y. (Eds) (1994). Entering the field of qualitative research. Sage Publications, Thousand Oaks, CA.
Gearing, R. E. (2004). Bracketing in research: a typology. Qualitative Health Research, Vol. 14, No. 10, December 2004, pp.1429-1452, Sage Publications, Newbury Park, CA.
Groenewald, T. (2004). A phenomenological research design illustrated. International Journal of Qualitative Methods, 3(1), Article 4 [accessed 21 Feb 2005] http://www.ualberta.ca/~iiqm/backissues/3_1/pdf/groenewald.pdf
Hill, B.V. (2001). What does it mean to call a school religious? Religious Education Journal of Australia, Vol.17, No. 2, p. 43.
Hogan, M. (1984). Public versus private schools Penguin, Melbourne, Vic.
Loader, D. (1997). The inner principal. The Falmer Press, London, UK.
Lukiv, D. (2004). Bracketing and Phenomenology. McNaughton Centre, Quesnel, Canada [accessed on 22 Feb 2005] http://asstudents.unco.edu/students/AE-Extra/2004/6/Art-1.html
Macquarie (1997). The Macquarie Dictionary. Macquarie University (3rd Ed) Griffen Press, Netley, SA.
McLaughlin, T.H. (1996). The Distinctiveness of Catholic Education. In T.H. McLaughlin, J. O'Keefe & B. O'Keffe (Eds), The Contemporary Catholic School: Context, Identity and Diversity, pp. 320-364. The Falmer Press, London, UK.
Miles, M. & Huberman, A. (1994). Qualitative data analysis, 2nd Ed, Sage Publications, Thousand Oaks, CA.
Moustakas, C. (1994). Phenomenological research methods. Sage Publications, Thousand Oaks, CA.
Myerhoff, B. & Ruby, J. (1992). A crack in the mirror: reflexive perspectives in anthropology. Quoted in Ahern, K.J. (1999). Pearls, pith and provocation: ten tips for reflexive bracketing, Qualitative Health Research, Vol., 9 No 3, May, 1999, pp. 407-411.
Praetz, H. (1982). Public policy and Catholic schools. Australian Education Review, No 17, ACER, Melbourne, Vic.
Patton, M. (1990). Qualitative evaluation and research methods, 2nd Ed, Sage Publications, Newbury Park, CA.
Ray, M. (1990). Phenomenological method for nursing research, quoted in Gearing, R.E. (2004, Bracketing in research: a typology. Qualitative Health Research, Vol. 14, No. 10, December 2004, pp.1429-1452, Sage Publications, Newbury Park, CA.
Porter, S. (1993). Nursing research conventions: objectivity or obfuscation? Journal of Advanced Nursing, 18.
Ratner, C. (2001). Analyzing cultural-psychological themes in narrative statements. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, Vol: 2, Article 3 [accessed 24 Feb 2004] http://www.qualitative-research.net/fqs-texte/3-01/3-01ratner-e.htm
Rieman, D. (1986). The essential structure of a caring interaction: doing phenomenology. Quoted in Creswell, J. (1998), Qualitative inquiry and research design: choosing among five traditions. Sage Publications, London, UK.
Rubin, H. and Rubin, I. (1995). Qualitative interviewing: a method to the madness. Sage Publications, Thousand Oaks, CA.
Schultz, S. (1994). Exploring the benefits of a subjective approach in qualitative nursing research. Journal of Advanced Nursing, 20, pp. 412-417.
Speigelberg, H. (1973). The phenomenological movement: An historical introduction. Quoted in Gearing R.E. (2004). Bracketing in research: a typology. Qualitative Health Research, Vol. 14, No. 10, December 2004, pp.1429-1452, Sage Publications, Newbury Park, CA.
Wengraf, T. (2001). Qualitative research interviews. Sage Publications, London, UK.
Weston, C., Gandell, T., Beachamp, J., McAlpine, L., Wiseman, C. & Beachamp, C. (2002). Analysing interview data: the development and evolution of a coding system. Qualitative Sociology, Sept 2001, Vol. 24 Issue 3, p. 381.
|Author: John Bednall PhD is the University of Notre Dame's nominee for the WAIER Postgraduate Award 2006 [Postgraduate Awards]
Please cite as: Bednall, J. (2006). Devising a phenomenological study within the qualitative paradigm. Proceedings Western Australian Institute for Educational Research Forum 2006.