Tag: fieldwork


FieldworkToday brought some interesting discussions on qualitative fieldwork, including on researcher roles and relations during data generation. First up was a COCAPP-A project meeting which included a conversation about observational methods in inpatient mental health settings. Second was a seminar led by Michael Coffey‘s PhD student Brian Mfula, drawing on his ongoing PhD experiences of ethnographic fieldwork centring on care planning and care coordination in forensic mental health care.

Brian shared his experiences of negotiating access, and of his reading and thinking about insider and outsider roles. This led to a wide-ranging talk amongst those present on fieldnotes and approaches to qualitative research (Grounded theory, anyone? Phenomenology? Or perhaps thematic analysis is more your thing?). We talked, too, about reflexivity, and knowing when (and how) to leave the field. Along the way this took us to the National Centre for Research Methods’ excellent Review Paper, How many qualitative interviews is enough?

Around ten years ago I contributed a chapter covering some of this territory to Davina Allen and Patricia Lyne’s edited book, The Reality of Nursing Research: Politics, Practices and Processes. Titled Data generation, this contrasted survey principles and practices in The All Wales Community Mental Health Nursing Stress Study with the ideas and methods in my (then-ongoing) ethnographic PhD, Health and Social Care for People with Severe Mental Health Problems. I wrote about decision-making, and the extent to which data are interactionally produced by researchers and participants together:

Whilst different strategies place different expectations and demands on nurse researchers, this chapter has also shown that – whatever approach is followed – data generation is always a purposeful activity demanding a reflexive stance. The principle of reflexivity underpins the idea that research always takes place in contexts, shaped to significant degree through an interaction between researcher and researched. The character of data produced in a study is moderated by aspects of the researcher’s personal biography and their interaction with research participants. This is a well-established principle in the social sciences. In nursing research, however, reflexive investigators have to give consideration not only to general biographical aspects such as age and gender, but also to their specific occupational backgrounds and practitioner experiences. A self-conscious, reflexive approach includes acknowledgement of the utility and the limitations of practitioner knowledge, and the implications of this for data production.

I’m now thinking that today’s seminar and discussions show how live these issues remain, and will ever remain so.

‘Psychiatric Ideologies and Institutions’: 50 years and counting

Happy new year. In the midst of a series of holiday period email exchanges Michael Coffey happened to mention that 2014 marks the 50th anniversary of the publication of Anselm Strauss and colleagues’ Psychiatric Ideologies and Institutions.

This is a fine book indeed, which during my time as a PhD student concerned with work and roles in mental health care was an absolute essential. In it, Strauss and his collaborators reported findings from prolonged and intensive fieldwork conducted in two North American psychiatric hospitals. Whilst today’s qualitative research reports will typically include lashings of direct data extracts, Psychiatric Ideologies and Institutions has little in the way of what Strauss et al referred to as ‘illustration and quotation’. Yet I never once recall, as a reader, doubting that Strauss and his team were truly there, participating in and recording everyday hospital life and its organisation.

It is at this descriptive level that the book initially works: as a meticulous account of the interplay between ideas, professions and practices in an area of health care which (both then, and to this day) happens to be particularly contested. One part of the dataset drawn on in the book comes from a questionnaire, designed to capture information about affiliations to particular treatment ideologies. From this nurses emerge as being ‘ideologically uncommitted’. In a later, detailed, section Strauss et al wrote of the problems faced by nurses in reconciling their managerial, administrative and therapeutic tasks and in answering the still-pertinent question:

…at the heart of her professional identity: What does therapeutic action toward patients actually involve for a psychiatric nurse?

My copy of Psychiatric Ideologies and Institutions is the edition published in 1981, for which a new introduction was added. In this, Strauss and his collaborators wrote of their original ambition to produce a book which was not only descriptive, no matter how detailed or accurate, but which was also theorised. It was the fieldwork and the findings reported most completely in this monograph that gave rise to the idea of the negotiated order. This is a sociological theory of importance which, in the decades following its introduction, went on to develop a life of its own. As Strauss et al wrote in their 1981 introduction, their original observation that theory might emerge from data represented a considerable methodological departure, more fully articulated at a later point with the introduction of grounded theory. Here, then, is a second way in which Psychiatric Ideologies and Institutions works, and remains of interest to people unconcerned with research into the world of mental health care: as an exemplar of how data and theory can dance together.

Today I’ve turned up this review of the book, which appeared in 1965 in the journal now called Psychiatric Services. In it the reviewer sums up with the recommendation that:

All in all, most professionals will find this book profitable to read, study and think about.

I concur, and commend this classic text to professionals and others alike. And as an aside, perhaps this short celebratory post can help persuade students (usually undergraduates, in my experience) that books and articles which happen to be more than five years old can still be worth reading.