Orientation Observation In-depth interviews Document analysis and semiology Conversation and discourse analysis Secondary Data Surveys Experiments Ethics Research outcomes



Social Research Glossary

About Researching the Real World



© Lee Harvey 2012–2020

Page updated 29 April, 2020

Citation reference: Harvey, L., 2012–2020, Researching the Real World, available at
All rights belong to author.


A Guide to Methodology

1. Basics

1.11 Reflexivity
Reflexivity refers to the deliberate self-evaluation by the researcher as to the impact that he or she is having on the research situation.

In view of the exploration in Section 1.6.1 of different epistemological positions, the discussion about the objectivity of social research and the problems of assuring validity, it is important for social researchers to be self-critical of their research.

Reflexivity includes:

(1) self-evaluation of the direct impact that the researcher has on the activities of the research subjects as a result of the data collection process;
(2) consideration of the indirect impact that occurs when the interim or final results are known by the research subjects;
(3) self-evaluation of the way in which the researcher explains, interprets or understands the research setting and constructs theories about what is being observed.

Mats Alvesson and Kaj Sködberg (2000, p. 5) provided a broad definition of reflexivity as a conscious and continuous attention to 'the way different kinds of linguistic, social, political and theoretical elements are woven together in the process of knowledge development, during which empirical material is constructed, interpreted and written'.

Reflexivity is often portrayed as a problem for 'qualitative' research, for observation studies in particular. However, all research, whatever the methodology, should be concerned about reflexivity.

All types of data collection have an impact on the research subjects, unless the research involves secret mechanical recording and even then the research is affected by where the recording devices are located.

Research subjects in all kinds of research are affected in some way by the data collection process. For example: they respond to the presence of a researcher in observational settings; they respond to data collection devices such as recorders in in-depth interviews; they respond to the unnatural surroundings of contrived experiments; they respond to non-verbal cues when constructing answers to scheduled interview questions; and they formulate responses to questionnaires in a way that reflects their understanding of the purpose of the questionnaire and, often, to minimise the work required in completing it.

Furthermore, all types of research involve a process of translating theoretical ideas into researchable concepts and a process of explanation, interpretation or understanding that results in the articulation of theories about the research setting, or at the very least a distillation of data to provide a description. This process of analysis and theorising is not a neutral process with natural self-evident outcomes but involves a series of decisions on the part of the researcher.

Natasha Mauthner and Andrea Doucet (2003, pp 414–15) stated:

there is an assumption built into many data analysis methods that the researcher, the method and the data are separate entities rather than reflexively interdependent and interconnected. Most methods continue to be presented as a series of neutral, mechanical and decontextualized procedures that are applied to the data and that take place in a social vacuum. Data analysis is described as 'a range of techniques for sorting, organizing and indexing qualitative data' (Mason, 1996: 7). The 'embodied', situated and subjective researcher carrying out the analysis is rendered invisible as are the interpersonal, social and institutional contexts. This positivistic model of the absent or neutral researcher is reinforced by computer aided programs for qualitative data analysis as 'the use of technology confers an air of scientific objectivity onto what remains a fundamentally subjective, interpretative process' (Mauthner and Doucet, 1998: 122).

It is, therefore, important that the researcher reflects on and makes transparent the research process, identifying where activities of the researcher and decisions made impact on the research outcomes (See CASE STUDY A method for reflexivity). This is reflexivity.

Nonetheless, despite it being relevant to all social research, the majority of the literature on reflexivity can be found in the discussion of ethnographic research. Ethnographers argue that it is important for researchers to develop a critical attitude towards their research practices, theoretical models and presuppositions. The following is the standard ethnographic approach to reflexivity, the views can be generalised to other research approaches. It is impossible to begin observational research without any preconceptions but the reflexive researcher will make an effort to confront those preconceptions and see what effect they have had on the research process. Field notes should be examined to see if the preconceptions are confronted. CASE STUDY Changing perceptions, from Blanche Geer (1964), is an example of how this is done.

Reflexivity also requires that researchers assess the contaminating effect of their presence and their research techniques on the nature and extent of the data collected. Crudely put, researchers must consider to what extent respondents were telling them what they wanted to hear or were acting in a manner to impress the researcher. Researchers must ask whether or not their presence inhibited respondents. Did the method of data collection, or even the format in which it was stored, restrict the kind of data being collected? The use of mechanical aids such as audio and video recorders also impacts on the research.

There has been little sustained attempt to develop a reflexive awareness of how recording methods affect the research process. Still less is it possible to assess how researchers and interviewees negotiated recording practices in the interview itself. (Lee, 2004, p. 882)

Furthermore, reflexivity requires that researchers critically reflect upon the theoretical structures they have drawn out of their analysis. Researchers are expected to reconceptualise their evidence and explore other possible models. They are expected to think laterally. Researchers should not just fit details into a preformed theoretical scheme but should try to see if alternative theoretical schemes provide the basis for a better understanding of the data. Do the observations reveal different meanings when viewed in a different way?

William I. Thomas (Thomas, 1923; Thomas and Thomas, 1928) was among the first to point to the need for reflexivity. He noted that 'the situations that men define as true, become true for them', suggesting that researchers can easily construct a version of reality that they assume to be correct without reflecting on alternative constructions. In similar vein, Robert Merton (1948, 1949) argued that a theoretical proposition made about a research setting can itself become self-fulfilling in as much as the researcher and the research subjects behave in a way to fulfil the prophecy of the research.

Since then, the disussion about the issue of reflexivity has encompassed methodological as well as methodic aspects. Mauthner and Doucet (2003, pp. 416–7) sum it up:

The 'problem of reflexivity' and the ways in which 'our subjectivity becomes entangled in the lives of others' (Denzin, 1997: 27) are issues which have concerned sociologists (Atkinson, 1992a; Denzin, 1989, 1995; Hammersley and Atkinson, 1983; Hobbs and May, 1993; Lather, 1991) and anthropologists (Clifford and Marcus, 1986; Geertz, 1980, 1988; Marcus and Fischer, 1986; Rosaldo, 1989) for at least 30 years, and philosophers for much longer (Quine, 1969; Rorty, 1979). The 'problem' arises through the recognition that as social researchers we are integral to the social world we study and as Denzin (1994: 503) points out, '[r]epresentation ... is always self-presentation ... the Other's presence is directly connected to the writer's self-presence in the text'. Feminist, postmodern, post-structural, hermeneutic, interpretive and critical discourses recognize that knowledge and understanding are contextually and historically grounded, as well as linguistically constituted. Feminists have been particularly vocal on this point (Grosz, 1995; Lather, 1991; Riley, 1988) and indeed reflexivity is one of the main themes in discussions of feminist research (DeVault, 1990; Fonow and Cook, 1991; Harding, 1992; Olesen, 1994).
The 'reflexive turn' in the social sciences has contributed towards demysti- fication and greater understanding of theoretically and empirically based knowledge construction processes. The partial, provisional and perspectival nature of knowledge claims is recognized. There is increased awareness that 'how knowledge is acquired, organized, and interpreted is relevant to what the claims are' (Altheide and Johnson, 1994: 486). The production of theory is described as a social activity, which is culturally, socially and historically embedded, thus resulting in 'situated knowledges' (Haraway, 1988). Attention has been drawn to the linguistic and rhetorical strategies used by social scien- tists in reporting their research and they are urged to reflect not only on their methods of data collection and analysis, but also on their methods of writing and reading (Atkinson, 1992b; Clifford and Marcus, 1986; Geertz, 1973). The supposedly 'neutral' status of texts has also been questioned, as different read- ers interpret texts in different ways depending on their social location and per- spectives (Denzin, 1994). As May (1998: 173) points out, this 'epistemology of reception' raises critical questions about 'how and under what circumstances social scientific knowledge is received, evaluated, and acted upon and under what circumstances'. The reflexive turn, and in particular postmodern and post- structuralist critiques, has thus created a sense of uncertainty and crisis as increasingly complex questions are raised concerning the status, validity, basis and authority of knowledge claims (Denzin, 1994, 1997; Geertz, 1988; Hollway, 1989; Richardson, 1997).


Next 1.12 Dichotomies