LSE - Small Logo
LSE - Small Logo

Blog Admin

March 13th, 2013

Overly Honest Social Science? The value of acknowledging bias, subjectivity and the messiness of research

4 comments | 2 shares

Estimated reading time: 5 minutes

Blog Admin

March 13th, 2013

Overly Honest Social Science? The value of acknowledging bias, subjectivity and the messiness of research

4 comments | 2 shares

Estimated reading time: 5 minutes

The popular Twitter hashtag #overlyhonestmethods reveals the widespread interest in methodological reflexivity. Jen Tarr reflects on the overt critique of scientific objectivity and argues good social scientific practice should be about acknowledging the weaknesses of methods to improve practice and to reassure novice researchers that real world research often is messy. Furthermore, we need more space in mainstream journals for over-honesty and the discussion of real methodological practices.

In early January, the hashtag #overlyhonestmethods began trending on Twitter. Started by neuroscience postdoc Dr. Leigh with a tweet about an undergraduate forgetting an experiment in the fridge, it quickly spread, as scientists from a variety of disciplines began saying the unsayable, breaking open the tidy reports constructed in the methods sections of most journal articles. Social scientists got in on the game too, with some of the best examples collected here. Responses from academic bloggers have been mainly positive (e.g. here) with only a few cautious voices suggesting that it devalues scientific practice or risks the careers of younger scholars.

We’ll never know how many of these apparent confessions are true. Some are certainly the stuff of nightmares for my methodological colleagues, the kind of practices against which we hope we’re inoculating our students: ‘I used quantitative methods because I assume the reader can’t parse correlation from causation’; ‘A qualitative data approach was chosen since our graduate stats course has permanently scared us away from numbers’; or ‘This method was used because my supervisor told me so when starting my PhD.’  My own worst fears about grounded theory are reflected in ‘Adopting a grounded theory approach, we collected all the data first and then made up a story about it afterwards.’  And I’ve seen too many examples of ‘Study site remains anonymous for confidentiality reasons but obviously it’s the University where I work.’

Most of us will laugh because these confessions sound all too plausible.  But reading them has also made me consider the implications this has for the impact and credibility of our work: not because I think that coming clean about actual methodological practices necessarily undermines the credibility of social science, but because I wonder about the differential distribution of over-honesty within different disciplines and methodological traditions.

Critiques of the supposed objectivity of science are not new.  Julius Roth flagged up some of these problems in the social sciences in ‘Hired Hand Research’ (1966) in The American Sociologist.  Science and Technology Studies practitioners like Latour and Woolgar (1979) have done the same in relation to laboratory science.  But methodological over-honesty has been slow to catch on in most fields.  However, there has been increasingly widespread acceptance of the confessional genre in qualitative research.  In the 1970s and 1980s when ethnographers began to be self-conscious about their writing practices and the inevitable role the researcher played in the production of the findings, it became more common for their monographs in particular to highlight these issues.  Paul Atkinson has written about the rhetorical tropes of the traditional ethnographic monograph: the bumbling antihero who, through making mistakes, comes gradually to greater knowledge of his or her ‘tribe’, a formula which sends the message that the research findings are reliable because ‘anyone could do what I did.’ Papers such as Judith Stacey’s ‘Can There Be a Feminist Ethnography?’ and Esther Newton’s ‘My Best Informant’s Dress: The Erotic Equation in Fieldwork’ have brought attention to less frequently identified challenges of interpersonal relationships formed in the field, which are as relevant to traditional monographs as they are to contemporary ethnography which has largely abandoned the quest for objectivity as such.

In part, ethnographers and other qualitative researchers have been forced to acknowledge these challenges in their research: the greater contact with participants required by most qualitative research methods inevitably leads to greater and messier involvement with them.  To produce qualitative social science, then, must involve some acknowledgement of these issues as part of writing up a good account of scientific practice.  The field of qualitative methods may be roughly split between people who think this kind of reflexive over-honesty has gone too far and that social research must cling to some semblance of scientific practice, and those who think it hasn’t gone far enough and that social researchers should give up on scientific rhetoric altogether.  But must acknowledging bias and subjectivity and the messiness of research fundamentally devalue that research as social science?

It shouldn’t.  Good social scientific practice should be about acknowledging the weaknesses of the methods used: not to reward sloppiness, but to highlight what really goes on; to reassure novice researchers that real world research often is messy; or to improve on current methods and find ways to do things better.  Better in this case does not necessarily mean less subjective, but it does mean more transparent, and usually more rigorous.  The publication of mistakes is a necessary part of this process.  Too often in both natural and social science, researchers face pressure to use established procedures simply to ensure results that work and appear ‘clean’ and uncontroversial, rather than experimenting with  approaches that might lead to better and more interesting data.  We need more space in mainstream journals for over-honesty and the discussion of real methodological practices.

Note: This article gives the views of the author, and not the position of the Impact of Social Science blog, nor of the London School of Economics. 

About the author

Dr Jen Tarr is a lecturer in research methodology in the LSE’s Methodology Institute, where she teaches qualitative methods.  She is particularly interested in new developments in visual and sensory methods, and the ethics and politics of social research. As a health sociologist, she has research interests in pain and injury and embodied somatic practices, and has published in journals such as Qualitative ResearchSociology of Health & Illness, and Ethnography. She has a PhD from Goldsmiths College, University of London.

Print Friendly, PDF & Email

About the author

Blog Admin

Posted In: Academic communication | Research ethics

4 Comments