Methods in Action Case Study
Using Participant Observation and Interviews to Study Middle-aged Gay Men, Ageing and Ageism
Author(s): Dr Paul Simpson, Dept of Sociology, University of Manchester.
Biography: Dr Paul Simpsonwas awarded his PhD in 2011 and lectures in Gender, Sexuality and Culture, the Sociology of Personal Life and qualitative research methods, including participant observation, interviews and biographical methods.
Methods Used: participant observation (PO) and semi-structured interviews (SSI).
Disciplines: Sociology; Social Anthropology; Social Gerontology.
Academic level: pitched at second and third year undergraduates and Masters degree candidates embarking on independent research projects leading to dissertations.
Keywords: Observation; interviews; purposive sampling; narrative/data generation instruments (interview and observation schedules); field notes; mixed qualitative methods; analysis; ethics and politics of research; plausibility and transferability of narratives.
Abstract
This case study addresses how participant observation (PO) sessions and semi-structured interviews (SSI) were combined in a PhD studyconcluded in 2011. Its principles could be applied to other mixed qualitative methods research projects. It considers the advantages of the methods used as well as how the methodological, theoretical and practical challenges that the methods presented were managed during the research. Specifically, the case illuminates the advantages of using these methods in tandem. It responds to some of the challenges of using these qualitative methods to produce accounts of everyday thought and practice that are both: plausible (rather than valid/accurate) i.e. credible to gay men and the academic community; and transferable (rather than replicable) i.e. could be heard in similar post-industrial cities with growing gay cultures or ‘scenes.’
Reason for the research
As a middle-aged gay man, I wanted to know why I was becoming more and more invisible when I went for a night out in Manchester’s ‘gay village’ (which has 36 bars and several nightclubs). Using insights from Sociology, it occurred to me that what seemed like private grief might actually have social causes – gay ageism. When I started to examine the literature on gay men and ageing, I soon realized that in the UK this was a neglected topic. Most existing studies were American and commonly assumed that ageing as a gay man was a downward slide towards rejection, isolation and misery. I set out to examine how being middle-aged (or an ageing gay self), was understood in a city with a highly developed gay culture. I also wanted to know what animated conflict or even lack of relations between younger and middle-aged/older gay men and what hopes middle-aged gay men could entertain for the future. For instance, what kind of relationships can they expect with gay and straight others?
Middle-aged gay men in Manchester, ageing and ageism: project overview
The research examined how middle-aged gay men (late thirties to pre-retirement age) living in Manchester manage growing older. It is based on accounts generated through 20 participant observation sessions conducted in Manchester’s ‘gay village’ and in-depth interviews with 27 gay-identified men. Through analysis of men’s body management practices (dress, grooming, diet, and exercise) and their relationships in various social contexts, the study examined the constraints on and choices around expression of midlife identity and study participants’ ways of relating. The study’s main theme concerned how midlife gay men in Manchester differentiate themselves from others–especially younger gay men.Whilst there are normative restrictions on expression of a midlife gay self and possibilities for interaction with younger gay men especially, middle-aged gay men can use self-worth and their political knowledges gained from life experience – what I call ‘ageing capital’ –to challenge these restrictions. The study also drew attention to the multidirectional power relations of ageism in (local) gay male culture. Midlife gay men are not just the targets of ageism from younger gay men. They distinguish themselves in ways that express ageism towards younger gay men, (some) peer-aged and old gay men. The study also challenges common stereotypes about midlife gay men: 1) that they are obsessed with the body, prolonging youth and maintaining sexual marketability; 2) that they are largely excluded from the ‘gay scene’; and 3) that they live isolated lives outside of family/kinship.
Learning Outcomes
By the end of the case you should be able to:
- think critically about the methodological advantages and challenges involved in using participant observation and interviewing and how to respond to some of the challenges the methods (and combining them) present to researchers;
- explain the value of observation and interview schedules for generating narratives/data that will provide answers to your research puzzle/questions;
- explain the value of purposive sampling of times, places (PO) and people (PO and SSI) to access diverse experiences of social phenomena;
- explain the value of a data analysisstrategy/framework that can illuminate the complexity of lived experience;
- explain how to address the ethics and politics of research – from conceptualization to dissemination – in a way that balances critical engagement with subjects’ accounts with recognition of their dignity and critical capacities.
Research design
I chose the two methods – participant observation and interviews – because they are better suited to capturing the rich detail, dynamic and often contradictory nature of personal narratives and social practices (Brewer 2000). Although in the ethnographic tradition, my research was not an ethnography that would, for example, have involved immersion in life of the gay village as a ‘bar fly,’ barman or a regular patron. It involved calculated retreat from ‘the field’ of research.Also, aspects of my use of participant observation resonate with a growing body of feminist and poststructuralist inspired auto-ethnographythat seeks to democratize the research process and recognizes the value of personal testimony of subjects and researcher. This approach emphasizes the epistemological problems with and political consequences of the notion of ‘objectivity’ (objective knowledge) and acknowledges the inevitable impact and value of the researcher’s subjectivity on the knowledge produced (Leatherby, Scott and Williams 2013).Sparkes (2002) urges that we become central characters in our writing and use our own embodied experiences as a valuable resource for inquiry. It also acknowledges the analytical and political value of recognizing subjectivity as multiform, of reflexivity on everyday (though hidden) gendering processes (Mykhalovskiy 1996) and in productive ways that escape self-indulgent, navel-gazing. Further, it links personal testimony with wider social, cultural and political issues (see Bertram 1998; Ettorre 2005). In my study, older gay men speak out to reveal personally felt but hidden, collective experiences of ageism and homophobia/heteronormativity and their testimony was used to examine the uses and limits of available theory.
Moreover, both methods are well suited to generating accounts (or ‘data’) that would allow examination of social processes rather than (factual) outcomes (Maxwell 1996) and that they could be productively combined to capture the multidimensional character of gay culture and men’s experiences of ageing. After all, the methods share a view of what it means to be a human subject. They rest on an assumption that actors have capacities to interpret the social world (epistemology) and see the social world as a product of human thought and interaction (ontology). Each method could yield different kinds of data that indicate different experiences of ageing and that each could compensate theoretically and practically for the limitations of the other. If, in-depth interviews could provide detailed, spoken narratives (Bryman 2004) that could be probed for meaning and would be difficult to tell/hear in village bars, PO could be used to generate detailed accounts ofdisplay of the body/self and interaction as it occurred in social space and point up some of the interesting contradictions between what actors say and what they actually do. This approach also enabled me to compare and contrast the different kinds of story and data. The plausibility or significance of events might be enhanced if stories/data from both methods point in a similar direction. In fact, using both methods in parallel created a ‘reflexive loop’ (Mason 2002) where observation data helped me to refine the questions being asked during interviews and interview data encouraged me to refine the foci/questions used in my observation schedule.
Observation roles and schedule
Because it was impossible to get the informed consent of people in the gay village to take part in the study, I largely adopted a covert observer role.Whilst I was generally more observer than participant (in noisy bars where prolonged conversation is difficult), in reality, this role became blurred and would shift when people would engage me in conversation and I became more participant than observer. Whatever the strengths or limits of these roles, taking time out from the field enabled me to ruminate on the meanings of what I had seen. Similarly, the line between covert and overt observation is not always clear-cut and became blurred when I shared with people who engaged with me my status and purpose for being there. On occasions, this would spark interest and lead to an on-the-spot, mini-interview where men would share their views on/experiences of ageing with me.
Moreover, when conducting two pilot observations, it soon became clear that the overwhelming number of events required an observation schedule with some minimal structure.With guidance from supervisors, I devised an observation schedule that consisted of a small column to record my thoughts about my own influence on the narratives being generated, a bigger column to note styles of dress, inter/action and a small column to record thoughts about concepts/theory that could be used to explain what was happening. This came in useful during the formal analysis and writing-up stage. The action column included brief memory joggers and my own shorthand (often recorded in toilet cubicles of various bars). These were elaborated into detailed description of events, feelings and insights as soon as possible away from the field. The field notes were later written up in descriptive detail, which included facial expressions, hairstyles, uses of the body, dance moves and detailed description of clothing. This instrument provided a conceptual/thematic structure i.e. a focus for each observation. The schedule was used to record seemingly commonplace or discrepant events that might suggest new lines of enquiry or ‘hidden’ experiences and the multiple realities of lived experience (Brewer 2000; Ettorre 2005). In particular, it helped generate stories of bodily expression and interaction that contest theories of gay male midlife and ageing as overly structurally determined and pointed up the more creative forms of self-expression and convivial forms of relating between gay men of similar and sometimes different generations.
Negotiating interviews and the interview schedule
If informed consent was nigh-on impossible in PO, it was vital if I were to act ethically in relation to men who might agree to be interviewed. Whenever contacted (on a project specific mobile ‘phone) in response to project publicity by prospective interviewees, I explained the purpose of the study, offered a brief description of myself and motives, a sketch of the kind of questions to be asked, approximate length of interview and ethical approach. If interested, an interviewee pack was despatched and followed up by an e-mail or telephone call to check willingness and negotiate a suitable time, date and venue. Interviews lasted between two and four hours, (averaging about two and a half hours) and were audio-recorded, with consent, for transcription. Time was spent building rapport and checking that consent was informed (explaining possible emotional risks). Informal ‘chat’ after interview questions and de-briefing about what the informant thought of the questions also yielded important ideas and data. So, by consent, it is advisable to keep the recorder running in this de-briefing phase of the interview.
I aimed for a conversational style of interview using a semi-structured interview schedule (SIS). This was designed to explore: responses to the signs of bodily ageing and body management with age (dress, grooming, exercise, diet); current patterns of close kinship and ways of relating compared to their twenties and thirties. It also asked for key biographical details: age; the respondent’s definition of his sexuality; occupation; qualifications; cultural, political and/or community pursuits; relationship status; and residence (whether renting, owner-occupier, living alone, sharing accommodation). These ‘background’ details are important to know because they could be key influences on thought and practice. The SIS combines order with flexibility: its structure lends a sense of coherence to the interview process whilst allowing for exploration of unexpected thematic, theoretical leads or significant ‘minority reports’ (Maxwell 1996). The more open-ended nature of the interview yielded accounts of experience that had not so much been anticipated, for example, concerning the detail of how men negotiate and even ‘gay’ the spaces they understand as ‘heterosexual.’
Criticisms of the methods
Both methods have been criticized by those working within more realist and positivist traditions of research for yielding theory based on anecdote. The data they produce whilst interesting are deemed invalid because they do not constitute objective, accurate or reliable (replicable/generalizable) social scientific knowledge. For instance, it was sometimes difficult when observing to ascertain who was ‘really’ middle-aged, gay or even male. Because it relied on my intuition about who fits or not within these designations, the method could have included ‘false positives’ or ‘category errors’ – younger and old gay men as well as heterosexual men. This methodological limitation might then lead to theoretical error, which risks misunderstanding gay male midlife as a distinct social location or set of experiences. But, this thinking overlooks that human beings are never perfect exemplars of the categories used to define them. Indeed, PO and SSI are much better suited than the survey method to deal with the ambiguity of categories (Robson 2002), which relies on the assumption that categories are fixed properties of the person and thus can miss the dynamism, complexities and ambivalences of identity and social experience.
Further, realist and positivist critics of qualitative methods argue that closer contact with the people whose actions/lives are being studied distorts or biases how actors think and their behaviour in their ‘natural’ habitats (Bryman 2004) But, I found that contact with the individuals observed and interviewed enabled me to understand their worlds in greater depth. PO offered ‘participation but not immersion and observation but not marginality’ (Burawoy 1991). Besides, withdrawal from the field allowed me to examine men’s experiences of ageing from a temporal and emotional distance (Spradley 1980). Even if we accept that ‘objectivity’ is valuable within social research, it can be understood in a different way in relation to qualitative methods. Rather than biasing the research, participant observation and interviewing brought me in touch with a different kind of objectivity in that it brought me closer to the ‘object’ of study – midlife gay male ageing – resulting in accounts that are more detailed, differentiated and plausible (Maxwell 1996). The test of ‘plausibility’ here is not whether stories are accurate, faithful or erroneous representations of gay male midlife/ageing but whether they are substantively recognizable by or credible to subjects and their (changeable) social positioning and the academic community. All knowledge is a position from somewhere rather than some impossible God’s eye view. Indeed, my study sought to escape the limitations of establishing the objective ‘facts’ of stories and was more concerned to examine the meanings of midlife gay men’s stories and illuminating the complex processes by which subjects produce often contradictory accounts of lived experience or fragmented realities (Bertram 1998).
Criticism has also been made that any data/accounts resulting from participant observation and other qualitative methods cannot qualify as (social) scientific knowledge because they are unreliable, i.e. they reflect subjective impressions of a few local contexts. But, both methods produced accounts that could transcend the environments in which they were generated (Brewer 2000). Accounts of ‘letting oneself go’ through disco dancing or signs of discomfort and shame in relation to the older body might be available in gay villages/scenes in comparable post-industrial cities like Birmingham, Glasgow or Leeds. Essentially, the criticisms above misapply the criteria of positivism onto an approach/methods that operate with very different notion of social worlds and how actors make sense of them (Kvale 1996). Researcher subjectivity and interaction with the subjects of research is not necessarily an impediment to but rather a resource for producing more finely tuned knowledge (Maxwell 1996).