Critical approaches in qualitative educational research: Representation and legitimation

Return to front page of resource

Introduction

Representation refers to the way that knowledge presented (or learnt or understood, or responded to – see epistemology page), is re-presented by the researcher. So long as knowledge is taken to be an external objective reality that can be described in factual, true statements, there no real problem about representation. However a problem arose in the second half of the twentieth century as a result of the growing influence of phenomenological qualitative research on the one hand, and action and process based inquiries on the other. This problem became known as the ‘crisis of representation’. At the same time and for the same reasons there was also a ‘crisis of legitimation’ about how research could be evaluated if there were no unchanging, measurable external reality against which to measure it. To put it another way, there is a question about how the validity of research should be judged. Denzin and Lincoln (2000) discuss all this in their influential model of ‘seven moments of qualitative research’ in which they divide the history of qualitative research into seven phases. This model is usefully summarised in Holt (2003).

Representation: What’s wrong with straight reports in plain prose that just stick to the facts?

All representation involves choices and judgements. The decision about how to edit and frame a report obviously involves judgement. But so does the decisions about which medium to use, and then within that, which genre to use. The medium may be text or it may be visual (pictures, diagrams), numeric (numbers, graphs), performances (theatrical, musical), or some combination of these. Each of these media supports a number of different genres. To take just one example, a text based report may be written as an orthodox research report, as a storied account of process, as the voices of the participants or as poetry. Orthodox research reports may be like reports in the sciences, or they may be like government policy documents. Storied accounts may be, implicitly, tales with a happy ending, or perhaps, a story of heroic failure. The voices may be presented without commentary or with academic or journalistic commentary. Poems can be sonnets, nursery rhymes or limericks. And so on and so on. Any account is influenced by the genres available to the teller: those available to the participants providing evidence, to the researcher and, of course, to the audience for the report. In other words, whether or not researchers acknowledge it, they are always already in their texts, in the judgements and choices they have made.

Furthermore, as Walshaw (2001) argues, both researcher and audience are embedded within relations of domination and subordination. So all the genres available, including the genre of ‘plain prose’ are already inscribed with these political relations (Walshaw, 2001, p.6/8):

These forces work through the text in ways that neither the writer nor the reader can fully anticipate or control, so that the text becomes inherently unstable, in flux, constituted by traces of other signs and symbolic statements. … Since there can never be a clear unambiguous statement of anything, including an intention or a meaning, the writer can never be sure how, at any given moment, the text will be interpreted. … It is in this way that knowledge is politically and historically constituted, ‘made’ by human communicative action.

Stronach and MacLure (1997) provide an example of how an apparently straightforward write-up of an life story interview with a head teacher, ‘Jack’, is structured by expectations of research paradigms and personal narratives – and by the perceptions of the writer, including their perceptions of gender. Two different write-ups are offered from the same interview, and compared with Jack’s own expectations about the representation of his own life story. This deconstruction does not, however, end up with a single suggestion about how Jack is to be represented. Rather it points not only to the limits of orthodox representations and their pitfalls, but also to the inherent openness of representation, so that it aims to ‘deny the reader that comfort of a shared ground with the author, foreground ambivalence and undermine the authority of their own assertions’ (p.57).

Some responses to the problem of representation

In response to the perceived problems of representation, researchers have experimented with a variety of other forms. One response has been the production of personal narratives. For instance, Marilyn Johnston’s (1997) tale of collaboration during a longitudinal study of school/university partnerships is an example of an account that uses narrative and preserves ambivalence as it aims to maintain tensions through dialogue, to acknowledge rather than wish away power relations and to demonstrate openness rather than the closure of an ending (happy or otherwise). Patti Lather’s (1997) article describes how she and her co-author, Chris Smithies, responded to the task of ‘telling stories that belong to others [the women with HIV/AIDS that they were working with] ‘in such a way that responds to the crisis of representation’ (p.286) but which ‘reaches towards a generally accessible public horizon’ (p.268). They created a ‘messy text’ including subtext commentary, poems, letters, speeches, emails and images.

Other responses have moved away from reporting stories to writing them or to non-textual representations such as film or drama. This may be termed a poetic turn in representation (Sparkes, Nilges, Swan and Dowling, 2003). Action research and self-study regularly include personal stories written by the researcher, alongside other data, which may be in non-textual forms (Weber and Mitchell, 2004; Mitchell, Weber and O’Reilly-Scanlon, 2005; Seeley and Reason, 2006). There has also been considerable interest in personal narratives written by the researcher as ‘autoethnography’. Campbell (2000) explains some reasons for not only fictionalising data but also for turning it into fiction. Clough (2002) discusses and exemplifies representing data in fiction, an approach discussed in de Freitas (2003), Sparkes (2003) and Wyatt (2007). Some researchers have turned to poetry (Rath, 2001; Saunders, 2003; 2006a; Spindler, 2008). Some have used visual representations or performance (Denzin, 2003a, 2003b; Kearney and Hyle, 2004; Weber and Mitchell, 2004; Griffiths, Windle and Simms, 2006)

Some of these alternative forms of data representation are the subject of an article by Eisner (1997) who assesses their ‘promise and perils’. The question of legitimation of these alternative forms continues to be relevant. The nature of truth, of evidence and of knowledge in alternative representations are widely discussed especially in articles themselves drawing on such representations (e.g. Campbell, 2000; Clough, 2002; Stanley, 2002; Denzin, 2003b; Sparkes et al., 2003; Holt, 2003; Saunders, 2006b; Griffiths and Macleod, 2008; Spindler, 2008).

Validity

The crisis of legitimation has not been resolved. Validity continues to be an issue for qualitative, critical research. Much of the debate is a response to the specialised use of the term in the natural sciences, and in that educational research which seeks generalisable kinds of knowledge (See epistemology page). The natural sciences have developed a specialised, technical vocabulary suitable for themselves. In this discourse, ‘validity’ determines whether the research measures that which it was intended to measure in order to determine how close to the truth the research results are.

One response has been post-modern playfulness. Suggestions for alternative understandings of validity include ‘rhyzomatic validity’ or ‘ironic validity’ (Patti Lather, 1993; 2003). Altheide and Johnson (2000) list ‘successor validity’, ‘catalytic validity’ and ‘transgressive validity’ among others. Another response has been to abandon the concept altogether as being bound up with the quest for certainty (Altheide and Johnson, 2000). In some discussions, ‘validity’ appears to have been equated with ‘quality’, as in the two linked articles by Heikkenen et al. (2007) and Feldman (2007), or even with truth itself (Hammersley, 1990). Other suggestions can be found in a useful article by Sparkes (2001). My own suggestion is to draw from the original understanding of the term (Griffiths and Macleod, 2008). We point out that ‘valid’ was not originally a word especially associated with measurement. Rather it comes from the Latin validus, meaning ‘strong’. So valid generally refers to the strength of evidence and arguments. The word retains this original meaning in ordinary speech. Only sometimes does it refer to measurement.

References

Altheide, David and Johnson, John (2000) Criteria for assessing interpretive validity in qualitative research. In Norman K. Denzin and Yvonna S. Lincoln (eds.) Collecting and Interpreting Qualitative Materials, Sage.

Campbell, Anne (2000) Fictionalising research data as a way of increasing teachers’ access to school-focused research Research in Education, 63 May

Clough, Peter (2002) Narratives and Fictions in Educational Research, Open University Press

de Freitas, Elizabeth (2003). Contested positions: How fiction informs empathic research, International Journal of Education and the Arts, 4 (7) .

Denzin Norman, K. (2003a) Performance Ethnography: Critical Pedagogy and the Politics of Culture, Sage

Denzin Norman, K. (2003b) Reading and writing performance, Qualitative Research 3 (2)

Denzin Norman, K. and Lincoln, Yvonna S. (2000) Handbook of Qualitative Research Thousand Oaks CA, Sage

Eisner, Elliot W. (1997) The promise and perils of alternative forms of data representation, Educational Researcher, 26 (6)

Feldman, A (2007) Validity and quality in action research Educational Action Research 15 (1)

Griffiths, Morwenna and Macleod, Gale (2008) Personal narratives and policy: never the twain? Journal of Philosophy of Education 42 (s1)

Griffiths, Morwenna, Windle, Joseph and Simms, Margaret (2006) ‘That’s what I am here for’: Images of working lives of academic and support staff. In D. Tidwell and L. Fitzgerald (eds.) Self-study and Diversity Springer

Hammersley, M. (1990) Reading ethnographic research: A critical guide, Longmans

Heikkinen, H.L.T., Huttunen, R. and Syrjälä, L. (2007) Action research as narrative: five principles for validation, Educational Action Research 15 (1)

Holt, Nicholas, L. (2003) Representation, legitimation and autoethnography: An autoethnographic writing story, International Journal of Qualitative Methods, 2 (1)

Johnston, Marilyn A. (1997) Contradictions in Collaboration: New Thinking on School/University Partnerships, Teachers’ College

Kearney, Kerri S. and Hyle, Adrienne, E. (2004) Drawing out emotions: the use of participant-produced drawings in qualitative inquiry, Qualitative Research, 4 (3)

Lather, Patti (1993) Fertile obsession: validity after poststructuralism, The Sociological Quarterly, 34 (4)

Lather, Patti (1997) Drawing the line at angels: working the ruins of feminist ethnography, Qualitative Studies in Education, 10 (3)

Lather, Patti (2003) Issues of validity in openly ideological research: between a rock and a soft place. In Lincoln, Yvonna S. and Denzin Norman, K. (eds.) Turning Points in Qualitative Research, Altamira

Mitchell, Claudia, Weber, Sandra and O’Reilly-Scanlon, Kathleen (2005) Just who do we think we are?  Methodologies for autobiography and self-study in teaching, RoutledgeFalmer

Rath, Jean (2001) Representing feminist educational research with/in the postmodern: stories of rape crisis training, Gender and Education, 13 (2)

Saunders, Lesley (2003) On flying, writing poetry and doing educational research, British Educational Research Journal, 29 (2)

Saunders, Lesley (2006a) ‘Something made in language’: the poet’s gift? Management Decision, 44 (4)

Saunders, Lesley (2006b) Teachers’ engagement in and with research: supporting integrity and creativity in teaching, Forum, 48 (2)

Seeley, Chris and Reason, Peter (2008) Expressions of energy: an epistemology of presentational knowing. In (eds.) Pranee Liamputtong and Jean Rumbold: Knowing Differently: Arts-Based & Collaborative Research, Nova Science

Sparkes, Andrew (2001) Myth 94: Qualitative Health Researchers will Agree about Validity, Qualitative Health Research 2001 (11)

Sparkes, Andrew (2003) Review Essay: Transforming qualitative data into art forms, Qualitative Research, 3 (3)

Sparkes, Andrew, Nilges, Lynda, Swan, Peter and Dowling, Fiona (2003) Poetic representations in sport and physical education: Insider perspectives, Sport, Education and Society, 8 (2)

Spindler, John (2008) Fictional writing, educational research and professional learning, International Journal of Research & Method in Education 31 (1)

Stanley, Liz (2002) Mourning becomes . . .: The work of feminism in the spaces between lives lived and lives written Women’s Studies International Forum, 25,(1)

Stronach, I. and MacLure, M. (1997) Jack in two boxes: a postmodern perspective on the transformation of persons into portraits, Chapter 2 of Ian Stronach and Maggie MacLure Educational Research Undone: The postmodern embrace Open University Press.

Walker, M. and Unterhalter, E. (2004) Knowledge, narrative and national reconciliation: storied reflections on the South African Truth and Reconciliation Commission, Discourse: Studies in the Cultural Politics of Education 25 (2)

Walshaw, M.A. (2001) The question of knowing others: Doing research in mathematics education, Philosophy of Mathematics Education Journal, 14 (1)

Weber, Sandra and Mitchell, Claudia (2004) Visual artistic modes of representation in self-study. In J. Loughran, M.L. Hamilton, V. LaBoskey and T. Russell (eds.) International Handbook of Self-Study of Teaching and Teacher Education Practices New York: Kluwer

Wyatt, Jonathan (2007) Research, narrative and fiction: Conference story The Qualitative Report12 (2)