Abstract
Institutional ethnography (IE) is being taken up by researchers across diverse disciplines, many who do not have a background in sociology and the antecedents and influences that underpin Dorothy Smith’s distinctive IE method. Novice IEers, who often work with advisors who have not studied or conducted an IE, are at risk of straying from IE’s core epistemology and ontology. This second of a two-volume set provides a broad overview to approaching analysis once the IE design and fieldwork are well under way. The purpose of two-volume series is to offer practical guidance and cautions that have been generated from my experiences of supervising graduate students and my involvement in reviewing and examining IE work that has gone “off track.” With a particular focus on the practicalities of conducting analysis, the paper includes examples of the application of IE’s theoretical framework with techniques for approaching and managing data: mapping, indexing, and building preliminary accounts/“analytic chunks.” I suggest these techniques are useful tactics to work with data and to refine the formulation of the research problematic(s) to be explicated.
What Is Already Known?
Analysis in institutional ethnography (IE) requires researchers to establish and maintain the distinctive “ontological shift” that is the hallmark of IE research. Novice IE researchers are often challenged by their tendency to import concepts and theories onto the data. A well-designed IE can be pulled seriously off track during analysis.
What This Paper Adds?
This paper contributes to the accessibility of practical advice about how to apply IE’s distinctive mode of inquiry. It draws on IE’s theoretical framework to provide suggestions for gathering and working with data. It is written for an audience of student researchers and their advisors from fields of practice that have not included training in IE nor a formal background in sociology, The goal is to illustrate, with examples, how IE conceptualizes social life for analysis, and to clarify how IE analysis differs from the conduct of qualitative research methodologies.
This sequel to Conducting analysis in Institutional Ethnography: Analytical work prior to data collection (Rankin, 2017) offers guidance in how to do institutional ethnography (IE) analysis in a way that that keeps it on track with its theoretical underpinnings. Developed by sociologist Dorothy Smith (1987, 1990a, 1990b, 1995, 1999, 2005, 2006, 2007), IE makes good sense to many people and is gaining popularity among scholars in professional disciplines who do not have a strong grounding in its intellectual influences and antecedents. 1 IE is conducted within assumptions about the social world and people’s lives that guide an examination of what is happening. Theory and method are combined into an inquiry that makes explicable everyday happenings that otherwise remain mysterious or misrepresented. The expressed purpose of IE is to generate potentially useful knowledge for people whose everyday activities are being organized against their own interests.
IE’s social ontology establishes credibility for the IE assumption that people enact social life and any one person’s activity “necessarily implies the presence and doings of others caught up in and participating in relations” that coordinate what is happening (Smith, 2005, p. 43). D. E. Smith’s (1987, 2005) term ruling relations expresses how this coordination “hook(s people) into the social relations of the economy and of ruling (institutions)” (p. 40) in ways that remain invisible until subjected to inquiry. Texts are principle instruments of ruling. Books, forms, computer fields, and many other “textual artifacts” (films, audio recordings, tickets, etc.) are replicated and circulated across time and location. The myriad texts that are ubiquitous in contemporary life have materiality. They coordinate people as they engage in what George Smith (1988) described as “text-work-text” processes (p. 17) that, in turn, generate “texts-in-action” (Smith & Turner, 2014, p. 8). Texts are important to IE’s social ontology. People’s use of texts gives ruling relations a material form that institutional ethnographers can use to investigate social organization. Ethnographic data offer instances of informants’ active and competent involvement with institutional texts and offer researchers insights into the everyday work of ruling.
Recognition of the real (not just theorized) social organization of people’s lives by “the strange forms of power that are at once present and absent in the everyday” (D. E. Smith, 2005, p. 41) guides IE researchers to empirically examine ruling relations implicated in the “problematic” (about which more is said throughout this paper) that establishes the research focus. To be opened up for analysis is the power activated within institutional complexes—the ruling practices that order contemporary Western societies. The ethnographic problematic locates the starting point of inquiry by anchoring it in people’s actual experience. Relying on the empirical forms of ruling relations, the research process tracks the institutional sites that govern people’s practices in local settings.
The work of the IE analyst is to conduct inquiries into ruling practices from the standpoint of actual people who occupy specific locations within the extended ruling regimes that coordinate everyday work. Beginning with ethnographic data about people’s particular positioning in the work of the regime, researchers can learn how these participants are active in, and subject to, the organized power of the institution. Smith’s important insight was that people’s knowledge from “being there” provides a grounded entry into the social organization of their experiences (knowledge) that otherwise may be misrepresented and/or hidden. People’s knowledge and experience hold the clues for tracking what actually happens in the processes of ruling. In this, there is an underlying assumption that we are all organized to participate in ruling relations; in the developed world, there is no one who is immune from their power.
The institutional ethnographer will encounter at every turn of the fieldwork, instances of hidden ruling relations that can be made visible. For example, in a wound clinic, a patient with a chronic foot ulcer knows that she should buy expensive orthotic shoes and avoid standing. However, her job as a sales clerk in women’s clothing requires her to stand for long periods. She cannot afford the shoes. Even if she could, she would not be allowed to wear them to work because they are not fashionable. The wound nurse records that the wound measurements are not improving. The nurse is becoming suspicious and troubled about how the patient is not complying with medical advice and begins to see the patient as someone who wastes precious health-care resources. In this social setting, the researcher would draw on her understanding of the standpoint as a way to direct data collection and analysis. The researcher’s choice to take either patients’ or nurses’ standpoint is a decision about which empirical location, within the workings of the clinic, will be examined. The chosen standpoint reveals different problems and different knowledge. Taking a specific standpoint provides a way to examine how knowledge works; whose knowledge counts. Without adopting a standpoint, a particular location within institutional practices, the researcher may be swayed by the apparent rationalities of dominant forms of knowledge—that most often arise in a standpoint of ruling. Thus, “captured” the researcher is unlikely to discover how different practices of knowledge work and how they are generated and activated in contradictory ways.
Within the conceptual framework of IE, taking a standpoint also establishes the materiality of the method. People occupy the standpoint, and it is people’s activities that link to other people across various institutional sites. Standpoint is an important methodological device that holds the researcher to describing material, tangible, tacit evidence that exposes ruling relations. IE’s insistence on establishing a standpoint reinforces whose interests the researcher examines and promotes within contested claims of knowledge. In the wound clinic example, the workings of “the clinic” are writ large in the work of both the patient and the nurse, but the tensions embedded in the workings of the clinic unfold differently in each person’s work. Taking a standpoint is a necessary methodological stance that directs what the researcher focuses on. It is the standpoint informants’ experiences and what those informants know and do that researchers must keep in sight as they talk to differently located people and interrogate the ruling rationality of the institutional texts and practices. Learning how the standpoint informants’ experiences are being coordinated is the purpose of the analysis.
This interest in coordination is the center of analytical attention in IE. Empirical discovery, description, and explication of the ways the lives of the people who occupy the standpoint position are being hooked into overlapping institutional relations of ruling are the focus of the method. In the wound clinic example, the ruling relations maybe connected by labor laws, disability insurance, shoe subsidies, the clinic appointment system, nurses’ workloads, and many other features of the clinic’s social organization.
The key point to be made in this introductory overview of some of the core characteristics of approaching analysis in IE is that IE constitutes a “radical departure” (D. E. Smith, 2005, p. 2) from mainstream sociology and from other qualitative approaches.
IE is an approach to inquiry that eschews both importing and developing theory about the issues being researched. However, IE is far from atheoretical. The approach employs a complex theoretical framework that has been generated through Smith’s intellectual engagement and integration of work done by scholars across a variety of disciplines. IE is complex and original, so much so that D. E. Smith (2005) characterizes the approach as a “paradigm shift” (p. 2). Most graduate students and their advisors who are undertaking IE research for the first time must rethink a great deal of their prior training about how to design a research project and how to gather and work with data.
In this paper, I use my own research and (with permission) 2 the studies of students I have worked with to contribute insight into how a researcher might advance an analysis that explores people’s experiences within “textually mediated social organization” (D. E. Smith, 1987). The advice (much of which is not new) expresses kernels of insight experienced in my own work into simple points and practical guidance.
Approaching Data Analytically: Practicalities and Pragmatics
The Research Problematics: The Overarching Guide
Analysis in IE is an iterative process that, in the early stages, is supported by the formulation of a research problematic. In my view, this is a critically important methodological tool that supports a coherent tracking and mapping (Campbell & Gregor, 2002) of ruling relations.
As with other research approaches, IE researchers often gather far more data than will actually be represented in the written dissertation or published papers. The apparently infinite institutional complexes evident in the data that are implicated in all the practices of a day-to-day life vie for the researcher’s attention. Research decisions about which features of the institutional matrices will be the focus of analysis are important junctures of the research process. While the researcher must sustain an interest in everything they know about the social organization of life in the standpoint, it is impossible to empirically track and explicate everything. To address this complexity, the IE analyst is provided with a methodological tool to support analysis—the research problematic.
In IE, a research problematic begins “with the everyday events in people’s lives, and in their problems of knowing—being told one thing, but in fact knowing otherwise on the basis of personal experience” (G. W. Smith, 1995, p. 21). The problematic is generated from the data, and it often rests on stories (accounts) that reveal troubles arising in (or conflicts between) authorized and experiential knowledge; whereby the tensions that standpoint informants know about and experience are either invisible or misrepresented within the authorized accounts (as was the case in the prior example of a patient’s and a nurse’s knowledge about living with a wound).
In my own doctoral work (Rankin, 2004), I established a standpoint in nurses’ working in hospitals and used accounts (from observations and interviews) that revealed differences among what nurses knew and experienced and how nurse managers, working at a distance from direct care, understood and knew about what was happening. I examined these contested views (the small problematics) and used them to inform a comprehensive problematic that operated as a way to build an IE explication of hospital reforms. As an example, my data collection involved me with a group of nurse activists who described serious things going wrong in their practice following the implementation of new policies to shorten patients’ time in the emergency and anesthetic recovery units. The shortened stays were initiatives that the nursing administrators described as improvements. There were other accounts from different things happening (i.e., patient satisfaction surveys that could not capture the hospital experience; words such as “holistic” and “quality” that had different meanings for differently located people) that revealed tensions. As an IE researcher, it was my job to identify such points of “disjuncture” (D. E. Smith, 1990b, pp. 83–104) and to elaborate these as the puzzles to be explicated. The research problematic I brought into view identified a contested definition of competent nursing practice that was linked to different understandings of “quality care.” It provided a way for me to focus on describing what got included and what got left out of nurses’ work, as it was being reformed to conform to the efficiency demands of restructured hospitals. On page 20 of my dissertation, I wrote: Throughout this dissertation, the problematic I outline here is used as a methodological strategy for discovery; it is integral to my research protocol. In the ensuing chapters of this dissertation I come back, time and again, to the stories I introduce here examining them “from the inside out” (Mueller, 1995, p. 106, cited by Rankin, 2004, p. 20).
Formulating a problematic is part of the process of analysis. It is a strategy that opens up the scaffold through which to analyze apparently disparate goings on that, at first, do not seem to be connected. For example, in my own doctoral work, my analysis and examination of data generated by a group of nurses who were banding together to challenge their administrators and later work analyzing a patient completing a satisfaction survey did not seem to have any coherence that linked them. They were empirical activities being carried out in different places at different times. The necessary material links that IE insists on were not immediately apparent. However, viewed within managerial ideas about quality of care and nursing competence (the iterative formulation of the problematic), I could delve more deeply to empirically describe (with evidence from the various sources of data) how nurses’ work was being represented in a way that did not fit what the nurses were saying or what the patient had actually experienced. The problematic is a methodological device that “direct(s) attention to a domain of possible questions, questions which have not yet been formulated, but which are implicit in the way the everyday world is organized” (Grahame, 1998, p. 350, italics added). Implicit or “latent” as D. E. Smith (1987) suggests, the problematic is often not apparent until you, the researcher, have spent some time working with the data.
Discovering the Latent: Looking the Traces of the Broad Institutional Relations in the Data
An opening approach to data and analysis is to develop sensitivity to the many, many clues into the institutional workings that the research settings offer for the inquiry. The data will be saturated with direct and indirect references to institutional practices such as meetings, appointments, schedules, policy, and rule-making. Moreover, informants’ talk will likely be infused with institutional language referencing people by their institutional position—principal, secretary, student, patient, resident, pharmacist, and so forth—terms that carry institutional traces. The researcher must pay attention to when the informant used specialized terms or acronyms that have been developed for the work or the setting. This sort of noticing is important in data collection and in the preliminary work with the data. It is a technique through which researchers train themselves to hear the traces of the institution’s otherwise taken-for-granted social organization in the informants’ verbal accounts and to understand that the institutional processes being talked about are the central interest for analysis.
Noticing the language and the institutional terms, during this very early engagement with data, when the interviews and observations are being developed into transcripts and detailed field notes, can be enhanced by the use of the comment feature in word processing software. I use this strategy to interact with the data using a fairly a “free-form” approach I elicit a conversation with the data. I highlight the traces of the institution, I write in the margins asking questions, expressing dismay, curiosity, or remarking on the unexpected.
Tensions and Contradictions for the Standpoint Informants: Querying Ruling Knowledge
The conceptual framework of IE informs how researchers look for tensions and contradictions. Research settings, like all social life, are the products of people working together. The IE researcher must “learn the setting” and also, often, learn how to think about the setting as IE analyst. In particular, researchers who are studying settings that they are familiar with or have worked in must examine prior knowledge, assumptions, and judgments. Often, captured by ruling ideas, it can be difficult for a researcher to see the latent tensions or contradictions that are being glossed over by the researcher’s prior assumptions and judgments.
Andrea Ingstrup (2014), a nurse who has worked in First Nations communities, conducted a study into the social organization of the mothering work of Jenny, a Canadian First Nations woman. Jenny lived on her hereditary land on a government “reserve” with her disabled toddler Crystal. She shared the home with her partner and his extended family. Ingstrup visited the home to interview Jenny and spent time learning about Jenny’s day-to-day mothering work, which was linked into multiple health and social services. Ingstrup accompanied Jenny and Crystal to appointments and (with permission from Jenny and consent from the professionals) she interviewed the professionals about their work. Some of the contradictions and tensions are included in Ingstrup’s first chapter of rich ethnographic data that describe what Jenny knows and what she is doing: Currently Jenny is trying to access income support for herself and also access child tax monies. For her income assistance Jenny requires picture identification to complete these applications. She does not possess any formal government identification…. (p. 43) Jenny expressed the sense of isolation and loneliness she experiences because she lacks access to readily available transportation. Beyond her personal sense of isolation there are external pressures from service providers both on and off reserve who expect Jenny and Crystal to attend appointments, meetings and programs.…Jenny says she is going to purchase a second-hand car with some child tax money to get herself back and forth to appointments. Jenny does not have a driver’s license. I asked her if she could get her license renewed and she laughed and said, “after I get my learner’s (license).” She then disclosed on two separate occasions she was charged with driving without a license and she cannot get her learner’s until she pays her fines. (p. 47)
As her first task, Ingstrup had to suspend her professional training and impressions. In order to begin to see things from Jenny’s standpoint, to clearly identify contradictions, and to commence an inquiry into the social organization of what was happening, Ingstrup had to overcome her own negative judgments about Crystal’s hygiene and set aside her taken-for-granted acceptance of the professional assessments of the nutrition team, surgical team, social worker, community health nurse, and so forth. Only when she looked at the contradictions from Jenny’s location—as a mother who was the subject of ruling relations—could Ingstrup begin to sort through the dense complexities of the data to begin to identify and describe some of the tensions related to the disparate interests active in Jenny’s work. For example, it only makes sense from a ruling standpoint/position that the Medical Transportation System penalized Jenny, who was mothering a disabled toddler under challenging conditions, if she was late for a pickup. 3 It was in querying the ruling relations and identifying tensions such as this one that challenged Ingstrup to rethink her own ideas about Jenny’s disorganization. It was an iterative process of IE analysis that allowed Ingstrup to firmly align herself with the standpoint of First Nations mothers such as Jenny and to use the data the way the IE method of inquiry directs.
Ruling Relations
In IE, you need to get from “contradictions,” “tensions,” “disquiet,” or “unease”—to interrogating what is going on. You need to train yourself to see how informants’ everyday life is being organized through an institution’s ruling practices. For Ingstrup, once she was firmly located in Jenny’s standpoint and had positioned Jenny as the “expert knower” of her life, Ingstrup began to analytically notice socially organized challenges in Jenny’s work that she might otherwise have glossed over within her own “ruling” ideas as a professional nurse. She began to link the contradictions and tensions that arose for Jenny into the authorized/professional knowledge about Jenny. She could describe and elaborate how Jenny and Crystal became the objects of a variety of institutional practices (the ruling relations) that were leading inexorably toward Crystal being taken away. In the fourth chapter of her thesis, Ingstrup (2014) wrote: It is inside the dailiness of the complexities I have described here that Jenny is threatened with Crystal’s apprehension in the care of Family Services. Several of the individuals I interviewed expressed in their professional opinion that in cases such as Jenny’s the risk that Crystal and Jenny’s new baby, soon to be born, will be apprehended is “very real.” (p. 99)
These early impressions supported Ingstrup to question how Family Services (and the complex of forms and documents that were inherent in social workers’ duties) coordinated barriers for Jenny that seemed unchallengeable—these were the myriad ruling relations that Ingstrup could choose to describe and examine. What the researcher learns about people’s work in the research setting informs the early insights that generate new questions. The emerging puzzles embedded in informants’ talk and the texts being examined will suggest how the setting and the work being done there is connected to activity going on elsewhere. I emphasize that although IE theorizes these connections as social and ruling relations, the IE researcher discovers how these relations are real and how they are organized in various “material” ways (texts and work processes).
Mapping, Indexing, and Writing Accounts
Mapping, indexing, and writing accounts are all useful analytical strategies that can be used separately and together. In particular, I have found that indexing and writing accounts “keeps the people in view” (Timothy Diamond, personal communication, March 23, 2011). Mapping can be approached in various ways. It supports the development of the analysis and is also a useful technique to bring visual coherence to findings. Indexing is a way to organize data into linked practices and happenings to support an analytic view into the institution. Writing an account entails selecting an instance of activity from the ethnographic data and describing how it is socially organized.
Mapping
In an interview with Bill (W. K.) Carroll (2010), Smith characterizes IE inquiry as “…a little bit like making a map. You can say, ‘This is how it’s put together. This is how things are going on.’ And people can use this” (p. 24). Later, in the same interview, Smith describes how she uses a map as a metaphor. She describes “…being in the malls in Toronto and you can find a map that says ‘You are here.’ And it is that kind of finger pointing off the text, into the world in which you stand, looking at the map or reading it” (p. 27). The map orients one to where one is in the world and endeavoring to “map” how something is happening is a way to approach the data—it is “indexical” to things happening. In mapping work, the ruling relations are first tracked from the local work of people into the work of other people. Then, the goal is to lay out a display of what is happening (the map), either in words or diagrams, that describes the features of the social practices and their respective material forms and relationships.
Ingstrup (2014) wrote two chapters of analysis that mapped out the ruling relations of Jenny’s “housing” work and her “program” work. Ingstrup relied on writing descriptions of specific instances in Jenny’s life and how these instances were coordinated within the texts and practices that Ingstrup had traced. She also developed handcrafted diagrams that she placed at the end of each of the chapters. The written accounts and the diagrams helped readers to grasp the complex processes and to see how they map into Jenny’s life—often in ways that worked against Jenny’s knowledge and her efforts. The written descriptions and the diagrams plotted out the various texts, the institutional applications, eligibility criteria, schedules, standardized assessment tools, and so forth. The ruling relations in Ingstrup’s mapping work linked into the Canada Indian Act, Treaty Seven, The Band Council, Housing Services, Alberta Health Services, Alberta Child and Family Welfare, Transport Services, Home Care Services, Nursing Services, a pediatrician, Nutrition Support Team, Parent Support Team, Family Wellness and Dental, and the Prenatal Care Team. The mapping work showed “Jenny is here”; the descriptions of Jenny and the diagrams of the services pointed to the world where Jenny stood and the ruling relations she was embroiled in. Reading Ingstrup’s mapping work expanded what could be known about what was happening in Jenny’s life, as she was subjected to the range of textual ruling practices that produced specific institutional responses. Ingstrup showed the “discursively organized practices where institutional power is expressed, made sense of, and enacted by participants” (Campbell & Gregor, 2002, p. 6).
Turner (2006) displays the texts she examined in her research in detailed schemata—analytic diagrams that depict extended work processes in complex textual systems involved in municipal land planning. These schematic maps supported her to see how the multiple texts embedded in sequences of action were a coherent “ruling relation” that coordinated the public consultation processes. Turner’s approach is an important contribution to analysis in IE. She describes how she “originally did the mapping work by hand” (p. 146) taping large sheets on a wall in her home and adding data as she discovered it. The schematic maps in Turner’s work became the core of analysis—enormously complex descriptions of what went on before and after the “public consultation” that was the focus of her research. Turner’s mapping strategy was a way to disentangle the complex economic, legal, bureaucratic, textual practices that were very complicated and had a great many interrelated parts. One of the core features of IE mapping is that it is “indexical”—a useful strategy to point to (or index) something going on in the place where it occurs.
Indexing
Congruent with the notion that a map is an indexical tool that can orient people to a physical location, Smith (cited by DeVault and McCoy, 2006, p. 39) suggests thinking about organizing data in IE in terms of “indexing” it. Indexing is a way of thinking that may help to avoid the common mistake of drifting toward “thematic analysis.” The practice of indexing is a way to discover linked practices going on in society. It is a strategy that can be used to stay grounded in IE’s core materiality as opposed to developing themes and categories that are abstracted from the data and that leave the particularities behind. Indexing is a tool that can be used to cross-reference across work processes, people, and settings. It is a useful strategy to begin the preliminary work with data.
One of my research projects was with a team of researchers examining how student nurses’ practicum evaluations are socially organized (Rankin, Malinsky, Tate, & Elena, 2010). We noted tensions (for students) that were hooked into weekly “reflections” the students wrote that were part of their teachers’ process to compose weekly “anecdotal notes.” The reflections and anecdotal notes were the documentary precursors to the formal midterm and final evaluations. In addition to the weekly writing, we discovered a variety of other work processes going on that were related to evaluation. We used these to index the data we had collected. For example, we noted that several faculty discussed “appeals.” We established an index heading “work focused on appeals”—we then indexed and subindexed all the work (all the purposeful activity generated among students and teachers related to evaluation/appeals) and the accompanying texts that we had discerned were linked to a student’s appeal (including the work of writing weekly reflections, anecdotal notes, and the related work of the semester coordinator and dean).
Indexing must be oriented to the materiality of the data. Indexing linked work activities is a way to organize one’s data around empirical happenings. A researcher could index all the work processes discoverable in the data that contributes to writing a report card, funding a nonprofit organization, developing global water policies, or securing a work visa. Similarly, texts that appear to be linked and nested could be indexed and/or mapped—in order to find their relationships and learn about when, where, and by whom they are activated.
I use indexing as a way to begin the analytical process when I have collected lots of interesting data—when I maybe floundering around, wondering what to do next! It is a way to engage with the data, to begin to sleuth out the disjunctures, and to describe what differently located people know and do. This helps with formulating a problematic. Currently, I am working on a project that is investigating nurses’ pain management work. The interest was sparked by ethnographic observations of hospitalized patients enduring long waits for pain relief. I am building an index of all the work processes involved in getting an opioid to a patient. The indexing strategy is uncovering extensive work practices and policies directed toward preventing illicit “diversion” of opioids by nurses. The “diversion prevention and accountability work” is in sharp contrast to the comparatively routine medication processes that require nurses to check the prescription and then to prepare and administer a medication. An early formulation of a problematic rests at the disjuncture of knowledge whereby administrators understand that the system is “working well.” This is the “authorized” view with evidence generated from interviews with the director of pharmacy and the chair of the regulatory and compliance committee. This contrasts with other evidence, the specific descriptions of nurses responding to a patient’s report of pain. My indexing of the practices is a way to begin to work with a great many texts and interviews, using a systematic data management approach that does not lose sight of workers and their practices.
Writing Accounts
In the research that investigated student nurses’ practice evaluations, as a separate data analysis strategy, and way to work with the data, we wrote two “analytic chunks” using data from two different students’ descriptions of writing reflections for their instructors about their practicum experiences. Our research team’s approach of writing accounts (analytic chunks) was done separately from the indexing work—as another way to engage team members with the data. The insights developed by writing the accounts exposed formerly concealed (implicit) work processes that we then put into the index.
Our preliminary analytical accounts began from the local, experiential expertise of the student standpoint informants. The accounts then focused on building up a description of people’s work and the institutional processes not visible from inside the experiences. The accounts were drafted to link back into the descriptions of the problems/tensions/frustrations/complaints that were embedded in students’ knowledge and experiences. The first account was developed about a student who described her work and her experience of being favorably evaluated. She attributed her success to having “learned how to bullshit” about her experiences in a way that was useful to and commendable by her teachers. The other account was based in the experience of a student who had been put on a “learning contract” for being disorganized. This student recounted how, as a result of the contract, she had ignored a patient’s incontinence in order to get to the clinical conference on time. We were interested in discovering whether and how these two very different accounts were linked into the same “generalizing relations”—the overarching texts and practices that ruled how evaluation proceeded.
The “accounts” included select quotes from each student’s interview. They also included data from the weekly notes, transcripts from the audio-recorded (formal) evaluation meetings between each student and teacher, and a preliminary analysis of, and excerpts from, the official “evaluation tool” (a checklist with comments) that had been generated for each student at midterm and end of term. The accounts were focused on developing coherent descriptions that began in actual instances of “being evaluated” but that expanded to include the dispersed talk, policies, schedules, and practices that were implicated in each student’s situation. The accounts captured the knowledge that informed what had happened, what each student knew about how to generate a satisfactory evaluation, and how faculty knew about what they were supposed to do in relation to assessing a student’s performance. Carefully detailing the students’ and the faculty’s work, we could see how many of the work processes were linked and we identified and described specific ruling relations that directed and coordinated what happened. The accounts we wrote were the preliminary work that began to describe the institutional processes that were being followed—the account of the institution. They helped us to consider how to formulate (and reformulate) the research problematic. The students’ experiences remained the central interest of the research. From that position, inside the ruling relations, the accounts illuminated the chafing everyday tensions embedded in the students’ efforts as they bumped into the ruling institutional processes.
Our engagement with the data provided a way for us to find useful analytic threads. A thread that drew our analytic interest was how we began to understand that the teachers were being organized to gather “evidence to fail.” The documents that the teachers were completing and the directions the teachers were given (at orientation, during faculty “student progress” meetings and even a description of a faculty member’s performance review) provided the empirical proof of this troubling social organization. If a teacher made a decision to fail a student, the failure had to be buttressed by statements that were structured by “behavioral indicators” that were categorized within “core competencies” and “domains of practice.” Worked up into authorized “facts” about a student, the teacher’s work was to “prove” a student’s incompetence. The work of gathering evidence to support a decision to fail included a documented record of a student doing (or more often not doing) something that was expected of them, deciding which “category/competency” the transgression represented and writing it up in the evaluation templates in order to “establish a pattern.” The index we had made was mined to make the links and join the dots in each of the accounts. We could show how teachers were required to produce detailed evidence in the event of a student appeal.
Through this analytical process, we were finding our way into the core analysis of all IEs—the discovery of the ruling relations/social organization of people’s lives. We had descriptions of things happening that provided new insights that supported us to generate problematics that focused our subsequent work with the data. One “arc” of a problematic we identified was how the appeals process, understood to be a fair and transparent recourse for students—in students’ interests—was actually vested in an institutional interest of winning the appeal and upholding the faculty decision to fail. The actual work being undertaken included pervasive practices that demonstrated adherence to something called “due process,” that we exposed as a set of ideological practices that ensured that a faculty member’s decision to fail would be “airtight.” Moreover, we could show how the time and resources that were devoted to “gathering evidence to fail” undermined teaching and learning during practicum experiences.
Using the Problematic (s) as a Methodological Tool to Establish Analytic Coherence
The “methodological disciplining” of a problematic supports the focused work of explicating the ruling relations of something. Without a clear sense of what is being explicated, the researcher is at risk of losing analytic focus and may drift toward an unwieldy conglomerate of ethnographic data that lacks coherence in regard to unraveling a puzzle in everyday life. At its heart, the work of noticing the small problematics that provide direction for a coherent arc of the larger analysis is simply learning to notice the occasions when the knowledge generated in the daily doing of work is subordinated by, or in tension with, other (abstract) knowledge that is used or supposed to be used to decide and to act. The problematic keeps the researcher focused on the everyday workings of the ruling relations operating in the standpoint informants’ experiences. Analysis is anchored in relation to the problematic(s).
It is often the case that a particular interaction (a single data excerpt) holds a great many traces of what is being problematized for inquiry. In writing the accounts, the researcher will likely discover rich pieces of data that ground the inquiry in the actualities of the lives of real people and provide a way to formulate problematics that lead to the discovery of connections—the ruling relations (texts, etc.) that open the institution to the unique investigative approach of the institutional ethnographer.
In the case of First Nation’s mother Jenny, researcher Ingstrup (2014) described how Jenny’s and her daughter Crystal’s lives were carved out into categories that were the various objects of the work of professionals in the institutional matrix. The professionals communicated with one another textually via referrals, assessments, and reports. These professional work processes resulted in Jenny’s life being (textually) broken “into pieces of health, mothering, development, economic resources and so forth” (p. 78). A central piece of Ingstrup’s data described the time when Crystal was treated in a hospital for Chicken Pox lesions that had become infected. The nurse at the hospital initiated a nursing referral for home care bathing. At the same time, Jenny was given teaching materials about vaccinations.
4
Duplicates of the vaccination directions were sent to the Parents’ Program Coordinator. Ingstrup used this textual data (and Jenny’s descriptions about what happened) to show how the texts generated activities that resulted in several new referrals, including a new dietary plan. Ingstrup noted how: Some of the new tasks that arose from the hospital visit required Jenny to make changes to her daily routines. These added to what she is already doing in her mothering work. Significantly they required Jenny to make time for the needs of the home care visitor. (pp. 87–88)
To reiterate, good analysis is guided by identifying a problematic arising in the everyday life of standpoint informants; the puzzle that the researcher is obliged to describe and explicate. Often a problematic can be formulated using only a snippet of data from a key interview or observation, a concrete something that happened. The situations described maybe unique and particular, but what matters is how the descriptions of people’s doings provide glimpses into the ruling relations being activated—those relations that have broader implications. For the informant, the “happening” may be ordinary, merely something they know must be done and that they know how to do. But the researcher brings IE’s theorization and assumptions about “the social,” the ruling relations, the institution, etc. to the data.
Problematics embedded in selected pieces of data provide a concrete place to start, they produce the preliminary stages of turning the apparent chaos of masses of data about everyday life into specific, socially organized instances of a something that can be taken up and explicated as IE inquiry. The researcher examines these entry points—descriptions of people and happenings—to determine the relations that exist and to discover other manifestations of those relations that were not previously apparent. Once you are “there” as IE analyst, those other manifestations turn up and make new sense to you.
Thus, the practical strategies I describe: mapping, indexing, and writing accounts are suggestions; ways to start working with the data to support the process of discovering as you go. They are a way to identify and follow your hunches and to begin to formulate problematic(s) that lead to discoveries. They are analytic strategies that are constantly framed by the IE mission of linking instances of things happening into the ruling relations that organize and coordinate those instances. They provide a resource for weaving the analysis together to show how the ruling relations work as generalizing practices that, your textual evidence will show, unfold in similar ways for variously located people across different times and in different situations. The IE researcher builds the coherent, empirical account that an IE paper, thesis, or dissertation must make.
A List of Cautions
The stimulus to write a practical guide for IE analysts came not only from my own work with graduate students but also from reviews I have done for journals and as an examiner when I find (too often) that work being characterized as IE has gone off track. This results in my decision to reject hours and hours of a researcher’s work. Most often, the analysis goes wrong because it abstracts from the data and/or makes broad assertions without showing empirically how people’s everyday problems are being socially organized. Sometimes the researcher uses Smith’s concepts the same way they would use Bourdieu (1984), Foucault (1970, 1984), Habermas (1981), or Marx and Engels (trans. 1976)—as a theory to be applied—that finds instances in the data that “match” the theory. In this way, they revert to a conventional ontology rather than actually using Smith’s social ontology. They superimpose IE terms onto the data rather than explicating the social organization of knowledge of everyday life.
In closing this paper, I have developed a list of the common problems I have run into. They include how: Researchers inappropriately try to combine IE with other conceptual frameworks or methods of inquiry. For example, IE combined with Mol’s (2002) use of actor network theory; Bourdieu’s (1977) practice theory or Andersen and Collins’ (2016) applications of intersectionality. While IE may be congruent with other critical analytical frameworks, IE’s social ontology demands an explication of the materialized social relations that coordinate what actually happens in the practices of people—that the researcher problematizes. IE is unique in this regard. While IE research may be paired with another approach to shed different lights on people’s problems, it should not be blended with other conceptual frameworks because (in all such work I have reviewed) this blending inevitably pulls explication off track. Researchers write a superb methods chapter that is an accurate description of the development of IE and the core tenets necessary for the ontological shift, standpoint, ruling relations, texts, and so forth. However, all too often, the clarity of the methods chapter does not translate into the actual research and analysis. Researchers have the words and the language of IE but have not shifted their thinking in line with the demands of the method and its core interest in social and ruling relations as empirical practices that are linked across time and geography. Researchers become absorbed in the descriptions of standpoint informants’ lives and begin to make the analysis about the standpoint informants instead of about the institution and a coherent description about how ruling and social relations coordinate what people in the standpoint know and do. Researchers become trapped in the local. Researchers become focused on showing us how somebody’s work is “disappeared,” “invisible,” or “subordinated.” They treat this as analysis. They do not show what is happening, how it is happening, and why it is consequential. Researchers write the account of the chronology of each discovery—the researcher’s journey through the research—rather than an account of the institution and how it works. Researchers leave readers to see for themselves what is in the data. Data excerpts are included but not explicitly unpicked as examples of something being socially organized. As well, researchers tell readers what we should see in the data rather than showing us how something is being organized in a way that readers can follow along. Researchers lose sight of explicating something happening. They write pages and pages of description but lose track of the task of explicating a problematic in the world. They depart from the main task, which is to conduct an inquiry that explicates how experiences (the rich, thick description) are organized to happen. Of course, this sort of writing is often how one gets to analysis, but it can also take people down a great many “rabbit holes.” My suggestions for beginning an analysis, for handling ethnographic data to discover what needs to be problematized, and so on should lead a researcher toward a productive inquiry that links local happenings into extralocal ruling relations. Researchers purport to be taking the side of the standpoint informants but maintain a patronizing attitude of wanting to “help” that is embedded in the ruling relations. They maintain their own “expert” stance and unwittingly sustain existing relations of ruling. For example, a professional nurse taking the standpoint of a social carer (unregulated care aide) may not be able to see the skilled expertise and knowledge that the carer brings to their daily decisions. Similarly, activists may be drawn into taking a moral stance that can generate a form of outrage that inhibits the capacity to actually discover what is happening as social organization and to make the necessary empirical links. Likewise, researchers’ own professional training, values, and beliefs keep them captured by ideas about what is good and right and how professional work ought to proceed. Researchers may be critical of the practices of the standpoint informants. These views and judgments can overrule what is in the data to be discovered. Institutional discourses can harness the researcher to the ruling relations and impede good analysis. Positioning the standpoint informants as the “expert knowers” of their situated work, genuinely listening and watching for their skilled expertise, and learning from them what they know about the smooth running of an everyday work day—as well as reflexively developing awareness of one’s own habitual responses is the way to overcome “institutional capture” (D. E. Smith, 2005).
A Note to Advisors
My final insights would be incomplete without noting the problems students can encounter when their advisor’s novice knowledge of IE generates questions and critiques that are not congruent with IE. Advisor’s ideas are often fully saturated in the taken-for-granted rightness (indeed necessity) to impose prior theory and language onto the research data and findings. A student’s team may inadvertently focus a student on preformulated, categorical ways of thinking (i.e., categories of race, class, and gender) that steer explication off course. About using such categorical approaches, D. E. Smith (2009) writes: (o)nce we attempt to unpack these categories as social relations, they become ambiguous. They arise in the organization of struggle against inequalities that people experience. But gender as relations between men and women is not separable from the actualities of the experiences of racial oppression or of the inequalities of class. Nor is race separable from class. This does not mean, of course, that inequalities, injustices, and oppressions do not differentiate; movements for change mobilize and focus on issues that are relevant to particular groups. But treating the categories as locating discrete phenomena of difference bypasses, indeed conceals, the social relations of inequality in which they are interwoven. (p. 80, italics added)
A student’s IE research maybe misguided by a well-intentioned academic team who pull the student’s attention away from an empirical analysis of the data and what it is showing. Academic advisors may require students to discuss their work using ideological knowledge to structure and support the findings rather contrasting the IE findings with how the social world is discursively constructed in prior research.
For example, I was methods advisor for a student examining physical activity in aged care homes. The members of the committee, and the student herself, were accustomed to thinking about “barriers and facilitators” as a way to conceptualize physical activity for institutionalized elders. Moreover, the committee unconsciously used institutional language (such as “activities of daily living”) that produced an institutional category that glossed over and left out critically important aspects of the everyday work of living in a care home. In early meetings, the advisory team tended to offer well-intentioned comments and questions that actually pulled the student away from using the data to explicate the social organization of things that were happening.
Advisors need to recognize the significant differences between IE and other methods and to appreciate that IE is so complex and original that D. E. Smith (2005) characterizes it as a “paradigm shift” (p. 2). IE is clearly not quantitative research, but it finds an uneasy fit with research approaches interested in studying “qualitative” phenomena. Most often, the various approaches that are characterized as qualitative have established techniques to abstract from data with explicit goals to develop theory, interpretations, or “meanings.” These approaches are decidedly different from the empirical/materiality of IE’s core ontology.
Conclusion
Ultimately, conducting analysis in IE is similar to other research insofar, as IE requires researchers to carefully manage and engage with data. In this paper, I have attempted to characterize the complexity of IE’s theoretical framework while at the same time providing practical examples and interpretations that are accessible for novice IE analysts who may not have a background in sociology. I have emphasized that the researcher interrogates the data for specific evidence (i.e., tensions, work processes, institutional language, and traces of discourse). Researchers go back and forth into data (and possibly into the field) to formulate a problematic that can be explicated. The data management approaches I have outlined provide a way to look for evidence that can support developing lines of analysis. The researcher uses the data (texts, talk, and descriptions of things going on 5 ) to construct institutional ethnographic accounts of how happenings are socially organized. The analysis is reflexive, iterative, political, and relentlessly empirical. The analysis develops as one thinks and writes. The thinking and writing illuminate nuanced practices that expose links into the institution that are not evident at the outset. It takes time and patience to discover the arc of the analytic threads and to develop a coherent structure for the developing findings. Even when an arc has been established and an outline developed, new findings will emerge and decisions must be made about whether and how the new discoveries can be integrated into the big account about how an “everyday problematic” (D. E. Smith, 1987) is socially organized.
In this paper, I have worked to illustrate “IE’s stable analytic processes” (M. Campbell, e-mail communication, December 2, 2016)—an application of the methodological framework that remains relatively invariable across different IE projects. I have offered practical advice about how to engage with data that is based on my own involvements in IE research that other IE researchers may find useful. The advice I offer is not intended to be a stepwise recipe for analysis. I anticipate that other IE researchers will have different things to say about how I approach IE as a researcher whose disciplinary training is not in sociology. Moreover, other researchers may also have practical tips about how to get from the “chaos” of early data collection and analysis to a coherent discovery of institutional practices and ruling relations that expand what informants can know about their everyday challenges. The paper was developed to contribute to the growing number of existing resources about how to conduct IE.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.