What Do We Mean By a “Hard-to-reach” Population? Legitimacy Versus Precarity as Barriers to Access

History of Prison Ethnography

Nowadays, U.S. prisons are a hard-to-access research site. As Simon (2000) lamented, “The great tradition of studies of prison social organization in the United States…has largely ceased…. Much of the research now is conducted from afar through journalism, official reports, and litigation discovery” (p. 289). Likewise, Wacquant (2002) famously declared, “The ethnography of the prison thus went into eclipse at the very moment when it was most urgently needed” (p. 385). In their annual review article on prison culture, Kreager and Kruttschnitt (2018) commented, “When compiling a review such as this, one is immediately struck by how little recent research has been conducted within U.S. prisons, a prerequisite for understanding inmate society in the era of mass incarceration” (p. 278).

This was not always the case. In the mid-twentieth century, researchers entered prisons with comparative ease, publishing a number of books on prison life (e.g., Giallombardo 1966; Heffernan 1972; Ward and Kassebaum 1965). Some prisons even welcomed sociologists inside their walls alongside psychologists and social workers (see Wacquant 2002). At the time, research focused on how individuals adapted to the uniquely punitive environment of a prison. Prison scholar Donald Clemmer ([1940] 1958) wrote about “prisonization,” the process by which an individual adopts “the folkways, mores, customs, and general culture of the penitentiary” (p. 270). Similarly, Gresham Sykes ([1958] 2007) called the prison a “small-scale society,” wherein “The student of human behavior can find many theoretical issues suddenly illuminated…where numerous features of the free world have been drastically changed” (p. xxxii). Erving Goffman ([1961] 2007), too, piqued academic interest in prison as a “total institution” that promoted conformity to rules and deindividualized roles (see Ellis 2021). Scholars seemed to construe prison as an experiment, observing how a confined community organizes itself and how individuals adapt to a closed environment.

A slew of books categorized emergent social groups within prisons. Typologies of prisoners were used to predict behavior. Some classifications relied on background characteristics including racial, religious, and political identities (Irwin 1970; Jacobs 1976). Other typologies relied on a person’s orientation toward prison rules: the extent to which an incarcerated individual accepted or rejected prison rules marked their place in the prison subculture (Garabedian 1963; Giallombardo 1966; Heffernan 1972; Kruttschnitt and Gartner 2005; Sykes [1958] 2007). Earlier studies of prison life described a “convict code” or a set of informal social rules that promoted collective solidarity among inmates and oppositional behavior toward prison authority (Hunt et al. 1993; McCorkle and Korn 1954). Over time, the growing role of drug use and gangs in prisons “eroded a more solidary inmate culture” (Crewe 2005:477) in exchange for smaller-group factions inside prisons (Skarbek 2014; Trammell 2012). Typologies specific to women’s prisons focused on romantic relationships and what was termed “pseudo-families” or fictive kinship behind bars (Giallombardo 1966; Heffernan 1972; Owen 1998; Ward and Kassebaum 1965). These early studies of women’s prisons placed undue emphasis on sexual relationships and family dynamics inside prisons, perhaps due to a scholarly lens that viewed women in domestic roles (see Kunzel 2008).

Later, during the prison boom of the 1970s and 1980s, the prison ethnography effectively vanished. Prisons closed their doors to researchers (see Eason 2017; Rhodes 2001; Wacquant 2002). University institutional review boards (IRBs) added protections for prisoners as a vulnerable population. Extra precautions ensured that researchers were studying prisoners per se and not prisoners as a conveniently testable population. Departments of Corrections created their own research boards to manage data access and to review the merits, scientific soundness, and feasibility of proposed studies.

As prison gates kept scholars out, studies of prison culture increasingly relied on secondary sources of data. Social scientists got creative, turning to prisons records, memoirs, historical archives, legal documents, and prison town community observations to round out their data alongside a restricted number of interviews with incarcerated persons (Berger 2014; Eason 2017; Reiter 2016; Skarbek 2014). The opacity and impenetrability of U.S. prisons are regrettable in an era of overcrowding, with a rise in life sentences and concomitant aging prison populations, amid stark racial disparities, and the decline of rehabilitative programming in favor of coercive control.

My Experience Getting Access: The Problem of Institutional Legitimacy

The turning tides of prison access made my own chances at entrée seem ill-fated. Foucault (1977) argued that prisons hide punishment from public view. While this imperviousness may be more palatable for society’s sensibilities, it fosters a detrimental “out of sight, out of mind” mentality. A select few scholars had successfully published ethnographic studies of jails and prisons in recent years (e.g., Johnson 2017; McCorkel 2013; Walker 2016). My specific research focus was on the role of religion inside prison, and I hoped to conduct ethnographic observations and interviews. Knowing that 2.1 million Americans sit behind bars on any given day (Kaeble and Cowhig 2018) in a moment of changing public sentiment and potential reform, the imperative was ever greater to interrogate the everyday realities of punishment doled out by the criminal justice system.

My access efforts were marked by opacity, uncertainty, and rejection. Upon revisiting the field notes documenting my entrée, I found that the major barrier was one of institutional legitimacy. Drawing from sociological and psychological scholarship (Suchman 1995; Tyler 2006), I define institutional legitimacy as an assessment that a given action is legible, appropriate, and worthy within an institution’s socially constructed set of norms and values. The action I sought was the ability to enter prison gates as a researcher. I navigated bureaucratic red tape over the course of nine months before achieving institutional legitimacy. This required that the prison institution view myself and my project as decipherable in a language understood by prison officials, appropriate within their norms and values, and worthy of institutional buy-in. As I describe my access experience below, I point out key moments where setbacks and rejections can be framed as instances of illegitimacy to the prison institution.

First, I proposed the project to my university’s IRB. This included a full review for Human Subjects Research by a committee that included a prisoner representative. My application included a supplemental form detailing the precautions I would take in the study design given prisoners’ status as a vulnerable population. I devoted a great deal of time and meticulous attention to my application. After the first round of review, the committee recommended 17 stipulations to address before resubmitting my application. I spent several hours on the phone with the IRB representative discussing the necessary changes. I carefully revised my submission based on the 17 required stipulations, then resubmitted my application for review. The IRB convened and approved my project, conditional on my submitting a letter demonstrating that the Department of Corrections (DOC) supported my entry.

Although the university IRB process was lengthy, my proposed project was institutionally legitimate from the outset. For one, the IRB receives human subjects research as part of its very function, meaning my proposal would be immediately legible. Moreover, I was deemed appropriate to the IRB because I was a PhD student at the same university. It was expected that students like me would propose projects whose worthiness is evaluated by the “study objectives,” “background,” and “risk/benefit assessment” articulated in the submission materials. My application was legitimate to the IRB because of my status as a student and because the project fit within an academic program. Although the university IRB scrutinized the proposed project, institutional legitimacy meant that their scrutiny yielded suggestions through official bureaucratic channels on how I could revise the proposal to gain eventual acceptance.

Meanwhile, I began to navigate the application process for the DOC research board in the state in which I hoped to conduct fieldwork. I mobilized my extended social network, through which I met someone who volunteered in the women’s prison. After describing my project, that person introduced me to Ms. Varlack (all names are pseudonyms), the DOC’s Director of Religious Programming, who was formerly the Director of the DOC Research Committee. I wanted to float my ideas by someone with a bird’s-eye level of expertise. Sitting in her office at the DOC, I found Ms. Varlack to be an intimidating administrator who spoke with a sober tone. She explained that it was important that “You don’t come in with a negative outlook, thinking that prison is only a bad thing. Of course, we want to learn about the good and the bad, but some people come in thinking prison doesn’t help anything.” Eyeing me with some skepticism, she nodded. “It doesn’t seem like you’re that way.” Ms. Varlack seemed to be assessing my legitimacy in that meeting rather than the project’s legitimacy. She had not asked for any paper documentation of my study, only a face-to-face meeting in which I described my study on “the role of religion in a women’s prison.” As I shifted in my chair, Ms. Varlack analyzed my appearance, demeanor, and tone. To come across as appropriate and worthy in her eyes, she relayed, I needed to demonstrate a measured approach to my proposed project. The politics of getting approved for research seemed to hinge on appearing academically agnostic—rather than critical—toward the subject of study, in this case, the prison system. As Gibson-Light and Seim (2020) reflected on their entrée into men’s prisons, “the researcher must often—at least partially—articulate the world as seen by penal authority. This usually means promising to keep a relative distance from the dominated and to remain obedient, and somewhat loyal, to the dominant” (p. 672).

During our meeting, Ms. Varlack explained that I needed to complete the DOC’s five-page research application, describing the purpose of the study, my methodology, and my proposed timeline. The application included a section on the “usefulness” of the research to the institution and a section asking me to list hypotheses. Although forming hypotheses was atypical for my grounded ethnographic approach, I obliged. Along with my application, the DOC requested a copy of my curriculum vitae, documentation of my dissertation advisor’s approval, copies of IRB-approved informed consent forms, copies of my interview guides, and a copy of my IRB approval letter. It struck me that institutional affiliation was a prerequisite for having my application considered. The DOC required these forms to ensure that the project was ostensibly appropriate: in this case, deemed scientifically sound by university-affiliated academics.

I mailed in my application. Three weeks later, I received a rejection letter. The language was brief and generic. My proposal was not approved. Discouraged by the lack of transparency in the rejection, I turned to my committee members for advice. They suggested that I call Ms. Caulfield, the current Director of DOC Research Committee, to ask follow-up questions on why my application had been rejected. I prepared a 10-second description of my project and wrote scripted answers for questions she might ask. My goal was to request the ability to revise and resubmit my application and to discern where my initial proposal went wrong. Whereas I now felt confident that the project appeared appropriate from a scholarly point of view, and that I appeared appropriate as a researcher backed by a university, I suspected that the legibility and worthiness of the project might be the roadblock.

Believing that the success of my dissertation hinged on this call, I mustered up the courage to dial a phone number provided online. No answer. I left a message with the office secretary for Ms. Caulfield. Anxiously awaiting a return call, I drafted a script of responses to questions she might ask. I printed out my script and carried a copy folded in my pocket everywhere I went for the next several days. Each time the phone rang, my heart started racing.

Days passed. I never received a call from Ms. Caulfield. Perhaps an e-mail would be more effective, I reasoned. I wrote of my wishes to revise the application, asking whether I could discuss possible changes over the phone. Six days elapsed before Ms. Caulfield responded to my e-mail. She suggested I write an addendum to my application and offered to “provide feedback over the phone…keeping in mind that any changes would still need approval by the Committee.” I was thrilled. Despite Ms. Caulfield’s offer to talk on the phone, the phone calls I made were fruitless. No one answered, and my voicemail went unreturned. I was stuck.

In the week that followed, I felt unmoored and uncertain. Telephone and e-mail communication was not working. At the suggestion of one of my mentors, I leaned into a gendered approach: I would show up to the DOC Research Headquarters with freshly baked cookies. It was December, and I could use the phrase “holiday cookies” to make the delivery sound like a reasonable thing to do. The plan was to appear friendly and nonthreatening while getting the information I needed on how to revise my application.

In retrospect, I suspect that this rather unusual plan only worked due to my positionality. I donned my most professional-looking dress and straightened my curly hair, hoping to come across as feminine, youthful, and professional. If I were a person of color, or appeared to be from a different social class background, this approach might have backfired terribly. I baked two batches of my crowd-pleasing oatmeal butterscotch chocolate chip cookies, the ingredients for which cost US$40. This financial setback felt visceral as a graduate student, but I was privileged to be able to afford the risk. I drove to the research office, with an approachable smile glued to my face, ready to pitch the project in person.

No one was expecting my visit. This fact must have raised alarm for the security guard, who asked a number of probing questions. Why did I want to speak to the director of research at the DOC? I responded with my prepared speech. I must have sounded convincing—but only to a degree. Ms. Caulfield came to meet me in the lobby. She did not invite me back to her office.

I offered her the cookies, delivering my prepared spiel: “I know I’ve been bugging you about my project and I wanted to come bring you these cookies and thank you—” Ms. Caulfield interrupted. “Have you talked to Henry in Facilities?” I was taken aback. I had never heard of Henry, much less known to contact him. Ms. Caulfield told me Henry was the person to talk to and that she would connect me to him over e-mail. Her response was so abrupt that it sounded dismissive, ushering me to yet another person who would never return my calls. Still, I expressed gratitude for her time and emphasized how important this project was to my schooling. I smiled and stretched out my arms to offer the tin of cookies.

Despite forgetting to mention that these were “holiday” cookies, Ms. Caulfield accepted them with a grin, offering a cookie to the security guard. When we said our goodbyes, I wished her a warm “Happy holidays.” She looked me straight in the eye and said, “Merry Christmas. I prefer that.” I chuckled sheepishly and returned a “Merry Christmas. Happy New Year.” Ms. Caulfield must not have guessed that I come from a Jewish faith tradition. Had I worn visible markers of another religious identity, her congeniality may have differed.

I felt sure that my appearance had everything to do with this productive (albeit hurried) encounter. As a petite, gender-conforming woman, standing 5’2” tall and looking younger than my 25 years at the time, these factors may have been disarming. Furthermore, they connoted shared identities of whiteness, educational attainment, and middle-class milieu. My effusive demeanor—quick to smile, with a nervous energy—may also have signaled alignment with the norms and values of the DOC rather than signaling a critical approach. In the same way that institutions like schools and prisons disadvantage working-class people and people of color (Flores 2016; Lareau 2011; Shedd 2015), they privilege middle-class people and whites. A different social class or racial background might have stymied access efforts. Getting the institutional knowledge needed to revise my application likely hinged on factors related to privilege.

Showing up to the DOC research office and putting a human face on my proposed project must have convinced Ms. Caulfield that I warranted further attention. An hour after the cookie delivery, she sent a friendly e-mail connecting me to Henry, sharing my telephone number and instructing him to call. The following afternoon, Henry called me to discuss the project. We spoke for more than half an hour, during which time Henry explained how I might revise my proposal to make it suitable for DOC approval. The majority of his suggested changes concerned the language I used to describe my goals. “Incarcerated women” should be referred to as “female inmates” or “female offenders.” Henry, Ms. Caulfield, and the other key prison officials I spoke to used words like “evidence-based practices” and “breaking the cycle of rearrest.” Quoting directly from conversations with these DOC employees: They suggested I should investigate whether religion is “linked with recidivism” or test a hypothesis that “those who truly buy in [to religion] don’t come back to crime.” I might look at the “efficacy of different [religious] programming,” one administrator recommended. I recalled that in our meeting, and Ms. Varlack insisted that my proposal needed to describe processes that were “observable.” As an ethnographer, I viewed social interactions as observable, yet these social interactions were not legible as observables among prison administrators, who favored practical, applied research. Observables of interest included prisoner behavior, infractions, psychological adjustments, and future reoffending. As I wrote in my field notes, “I need to be…prepared to speak their language of recovery, behavior, and recidivism.” Henry’s suggestions tailored the legibility and worthiness of my proposal to institutional gatekeepers’ norms and values.

Importantly, in establishing a proposed study’s credibility in the eyes of institutional gatekeepers, it is crucial to think through implications for what sort of research ultimately gets conducted. Gatekeepers endorse research that is legible to them, excluding critical research that mobilizes different language and goals. This is as true for prisons as it is for other closed-off institutions. In their study of global elite workers at the World Trade Organization, Conti and O’Neill (2007) reflected on the necessity of “[s]trategic management of [the researchers’] political identity,” as “it enabled the interviews to proceed without provoking a defensive posture from the informants” (p. 77). Perceptions of political and ideological orientations may dictate who gains access and who does not, thereby shaping the production of knowledge around dominant institutions such that the most critical scholarship is selected out.

At the time, I believed my revisions were strategic and legible. Despite this, nearly two months later, I received another rejection letter. This time, the letter suggested I contact the prison warden directly, as “It was agreed that this project does not meet the interests of the [DOC].” I panicked again. Were “interests” of the DOC akin to “worthiness?” I did not understand why the new language I used led to a second rejection.

Ms. Caulfield agreed to connect me to the warden over e-mail. Thankfully, her message contained no mention of “not meet[ing] the interests of the [DOC].” Although the proposal was rejected a second time, it was clear that the research board assigned greater legitimacy to my project in that they were willing to facilitate a conversation with the warden. To my surprise, the warden agreed to talk to me. During our phone call, the warden invited me to meet in person to discuss the project on-site at the prison.

I drafted some talking points. A friend suggested I focus on how the research might benefit the institution to emphasize the worthiness of the project. On the day of our meeting, the warden bowed out, delegating the assistant warden and the chaplain to meet with me instead. When I arrived—nerves at an all-time high—the assistant warden was holding a hard copy of my research application. She inspected the application point by point, asking a number of clarifying questions. She said she needed to peruse the proposal to make sure all the “i’s are dotted and t’s are crossed.” Shaking her head, she warned, “We don’t want to get into trouble.” The assistant warden spoke of what “we” needed to do rather than what “you” needed to do to ensure the project’s appropriateness. The hierarchical nature of authority in the DOC was working in my favor (see also Brayne 2020). Apparent institutional buy-in from above shifted the tenor of the meeting from whether my project was suitable to how to implement the proposed study suitably.

After the meeting, I was granted permission for preliminary access for eight weeks of observations. For the next two months, it was my job to get to know the people who lived and worked in the prison. I sent handwritten thank-you notes to everyone involved in approving my entrée. Converting my eight-week access into a full year involved making my research goals legitimate to not only the prison institution but to specific institutional actors. For one, because my focus was on religion in prison, I established rapport with the chaplain. I initially offered to help with office tasks for a few hours once per week, but she seemed reluctant to take me up on it. I soon learned that walking into her office and asking “What can I do for you today?” worked better than “When might you need my help?” or “Let me know what I can do to help you in your office.” On the final day of my preliminary eight weeks, the chaplain agreed that I could continue my observations if the assistant warden approved. “The ladies have loved having you,” she explained. Building relationships with incarcerated women was part of the ethnographic process—not a strategic maneuver to prolong access—but looking back, it seems to have been essential. Access to prison for research required far more than filling out paperwork—it required a substantial amount of face-to-face interaction. It depended on equal measures of persistence, a disposition deemed nonthreatening, and the approval of those who held the keys.

None of the factors that made my project appealing to my dissertation committee were the same factors that facilitated my entrée. I rephrased my research questions to suggest practical outcomes and applications of the findings. So often in academia, scholarship is incremental and peppered with caveats. We chip away at research questions little by little; such is the nature of science. The DOC, by contrast, sought larger claims about the implications of a study, including measures of effectiveness of a given intervention without as much hedging. In terms of language, I referred to those who live and work in prisons using the labels that prison officials use, thereby conferring legitimacy to the institution. In the social sciences, we often design our projects to expose enduring inequalities, and our language mirrors that. In contrast, to secure access to the prison, the DOC needed to endorse my project as roughly aligned with (or at least not contradictory to) institutional values and goals. Finally, my role as a researcher was on trial. I found that face-to-face access efforts were more successful than bureaucratic channels to appear legible, appropriate, and worthy as a person. A congenial, unassuming demeanor made gatekeepers seem more willing to grant me permission. Institutional legitimacy as a barrier to access meant that the researcher is scrutinized as much as the research design.

History of Prisoner Reentry Studies

Reentry is increasingly viewed as an urgent topic on the ramifications of criminal justice contact, as 95 percent of those who serve time in U.S. state prisons will return home (Hughes and Wilson 2020). Returning citizens face obstacles at every turn, from family reunification to employment to housing. Serving time strains relationships with partners and children (Roberts 2004; Wildeman, Schnittker, and Turney 2012). The mark of a criminal record hinders employment opportunities (Fader 2013; Pager 2003) and college admissions (Stewart and Uggen 2019), compounding disadvantage alongside returning citizens’ exclusion from public housing, welfare assistance, voting rights, financial aid, parental custody, and even a driver’s license in many states (Allard 2002; Harding, Morenoff, and Herbert 2013; Mauer and Chesney-Lind 2002; Petersilia 2003; Uggen, Manza, and Thompson 2006). Analysis of reintegration and the collateral consequences of incarceration are central to the contemporary landscape of punishment and inequality.

Formerly incarcerated people are considered a hard-to-reach population. For one, recidivism rates are high, as the path to desistance is more “zigzagging” than linear (Laub and Sampson 2003; Maruna 2001; Panfil 2020). Nearly one in five returning citizens is reincarcerated within one year, and two thirds of returning citizens are rearrested within three years (Langan and Levin 2002; Western 2018). Individuals are enmeshed in a system of “wraparound incarceration” (Flores 2016; Rios 2011) within an ever-widening net of “shadow carceral institutions” such as schools, religious organizations, and civil courts (Beckett and Murakawa 2012; Ellis 2020; Guzman 2020). While offering social service provisions, reentry and treatment programs also sanction and surveil (Prior 2020), volleying individuals from prisons to halfway houses and back again (Huebner, DeJong, and Cobbina 2010; Richie 2001). This nonlinearity makes it difficult to study reentry as such. Furthermore, reentry research is expensive (Bushway, Stoll, and Weiman 2007) and retention rates are low (Nelson, Deess, and Allen 1999; Visher, Debus-Sherrill, and Yahner 2010). Transient lifestyles, frequent contact with criminal justice actors, and limited postrelease support systems make it difficult to identify, contact, and meet with the population of interest (Western 2018).

To access this hard-to-reach population, two methods of recruitment pervade the reentry literature. The first access strategy involves recruitment through institutions. Mirroring the institutional gatekeeper considerations described above, this tried-and-true strategy draws on a convenience sample from an organization that serves the population in question (e.g., Leverentz 2014; Martin 2018; Mollborn and Jacobs 2012; Prior 2020). For instance, Leverentz (2014) interviewed women from one residential reentry program, gaining permission to conduct interviews inside the facility, a familiar and convenient location for participants. One potential limitation of this method is the lack of privacy within institutions designed to surveil (Fahmy et al. 2019). Another potential limitation is the exclusion of key segments of the population in question—often the most vulnerable (Hagan and McCarthy 1998; Sweet 2019). Nevertheless, this can be an effective method through which institutional buy-in facilitates participants’ buy-in.

The second access strategy extends beyond institutional recruitment through reentry programs, recruiting from penal institutions then maintaining contact postrelease (Fader 2013; Flores 2016), often involving a research team (Fahmy et al. 2019; Farrall et al. 2016). For example, Western’s (2018) study retained 91 percent of their formerly incarcerated sample using a research team, conducting regular phone check-ins with subsequent e-mails and letters. The team maintained a list of secondary contacts, including friends and family, if efforts to reach the participant failed. Secondary contacts were often the product of a gendered division of labor, in which mothers and female partners of formerly incarcerated men offered stability for their loved ones. Finally, they worked with justice agencies and developed community contacts to locate participants. This feat of access and retention required an impressive amount of coordination and persistence, benefiting from a large, well-funded research team. What are the takeaways for solo researchers conducting this work—for smaller teams and less financial support? Moreover, what if women are the main focus of the study, lacking the gendered safety net that men rely on?

Studying returning citizens, especially women, is a challenge. Criminologists have demonstrated that women’s “pathways” to punishment differ from men’s, especially related to high rates of trauma, addiction, and abuse (Salisbury and Van Voorhis 2009). These pathways to incarceration also operate as gendered roadblocks to reintegration. Women face distinct hardships related to employment and parenting post-release (Bachman et al. 2016; Leverentz 2014), particularly given the symbiotic relationship between the criminal justice system and the foster care system (Roberts 2012). Addiction plays a central role in women’s persistence in criminal offending (Leverentz 2014). Social stigma levies a harsh penalty on women who have experienced incarceration, straining an already-limited social safety net (Dodge and Pogrebin 2001). Recruiting for a study on women’s reentry erects additional barriers among a hard-to-read population of individuals returning home from prison.

My Experience Getting Access: The Problem of Structural Precarity

Before entering the field, I expected that female-identified returning citizens would be a hard-to-reach population. Armed with the strategies gleaned from my earlier prison study, I contacted women’s reentry organizations in St. Louis, Missouri, trying to present myself and my project legitimate from the start. A colleague who researched similar organizations in the past introduced me to those in charge. Thanks to my institutional moniker and professional referral, I expected to appear legible, appropriate, and worthy. Believing that recruiting from reentry programs would be the most direct approach, I called and e-mailed a number of residential and nonresidential reentry programs, requesting to meet in person.

Meetings with the full-time staff at these reentry organizations proved straightforward: I arranged a meeting at their offices, sometimes with my graduate student RA, Amanda, in tow. We prepared photocopies of the IRB approval, recruitment instruments, interview guides, and interviewee payment forms and tucked them into brand new navy blue folders. I felt confident that these “interview packets” fostered a sheen of preparedness and legibility. In another bid for legitimacy, I tried to match the business casual attire I expected from reentry staff and social workers: a colorful, loose-fitting top, black slacks, and ballet flats. As I wrote in my field notes, the ideal clothing for the occasion was a “wholesome Midwestern Ann Taylor sweater.” At each of these meetings, staff seemed amenable to my research goals and the cooperation I sought from them as institutional gatekeepers.

Fortunate to have received grant support, we could offer $25 interviewee incentives. Interviewees were paid in cash, except in the case of one reentry program, whose director required that we offer $25 grocery store gift cards instead, which could not be used to purchase alcohol or tobacco. As she explained, “In the past, when some women have gotten cash, they leave their program right away,” implying that a cash incentive might interfere with women’s sobriety. Despite my trepidation at policing participants’ consumption, I complied with this request in hopes that it would maintain institutional assessment of the study as appropriate.

Yet, access was difficult despite the institutional legitimacy achieved with this project. Little did I anticipate nearly two months’ delay before the first interview would occur and nine months before the target of 48 interviews would be completed. I came to realize this was due to the barrier of structural precarity, defined as respondents’ everyday vulnerabilities as a function of their structural position. Per Fine et al. (2018), precarity, or “structural dispossession,” is the result of the neoliberal conditions that “undermine economic opportunities, threaten living and learning settings and heighten police surveillance and violence…[leaving marginalized populations] more vulnerable to the unnerving predictability of impending disaster” (p. 611). Normative access strategies must be questioned and adapted to the realities of daily life among research participants. Importantly in navigating structural precarity, no researcher is entitled to participants’ time and candor. Instead, the onus is on the researcher to render the study inclusive and feasible.

In the reentry study, the barrier of structural precarity manifested as follows. After reentry programs agreed to post flyers in their hallways and advertise the study to their clients, I waited for phone calls and e-mails to roll in. It took a few weeks before the flyers were posted, after which I received a handful of calls. A number of prospective interviewees said they were interested in the $25 incentive, especially with Christmas around the corner. Unfortunately, the timing was such that I received these calls while out of town at a professional conference. I answered the phone each time and told the prospective interviewee that I would call back the following week to schedule an interview. That was my first mistake. After returning from the conference just a few days later, some calls went unanswered, while others reached the wrong person. Several prospective interviewees had borrowed someone else’s phone to inquire about participation. The cell phone owners did not recall prospective interviewee’s names nor their whereabouts. Our first lesson was that telephone access was unstable. When interested participants called, we had to schedule an interview immediately.

There was a lull in recruitment. The phone stopped ringing. Winter break took Amanda and me out of our research site and back to our respective hometowns. We started again in early January. With the holidays behind us, the interviewee incentive intrigued less immediacy. The flyers were outdated, and reentry staff stopped mentioning our study. Amanda reported feeling disillusioned, calling the process “frustrating” and “an access nightmare.” When either of us missed a recruitment phone call while teaching or in a meeting, we would call back only to reach a full mailbox or a phone ringing off the hook. Farrall et al. (2016) argued that research teams improve chances of continued contact for this reason. We learned that successful recruitment meant fewer missed calls.

We continued with institutional efforts, calling reentry programs and asking them to circulate flyers again. A few phone calls trickled in. One such call came during a meeting with Amanda. I answered on speakerphone, and Georgia, a prospective interviewee, was on the line. Her voice sounded stilted and unsettling. I arranged an interview for the very next day. As soon as I hung up the phone, a weight hung over the room. “Did Georgia sound high to you?” Amanda asked. I pursed my lips. When Amanda arrived to interview Georgia the following day at the agreed-upon public location, no one showed. Despite repeated efforts at contact, that interview never happened. We knew that addiction might be a challenge faced by our research participants (see also Fahmy et al. 2019). Indeed, as the interviews progressed, we heard about the extent to which substance abuse was linked with carceral contact and the ways in which recovery and reentry were connected. For example, Marcia, a woman in her mid-50s who currently lives in a residential reentry facility, described the revolving door of incarceration related to her addiction:

I came home in 1991…started using. Using was always the most important thing…I’d repeat that same behavior and be locked up within a year…. Again was home less than 18 months…and just repeating the same behavior again. When I went back, went to the long term treatment and still came home and used…. Was still getting high, stealing the whole time…. And then I went back [to prison] for six years and nine months…. I had a new case within six months of getting out on parole…. And ended up using again…. I received a 14-year sentence…. Again I came home and used.

Other times, interviews were canceled due to last-minute scheduling changes related to community supervision. Among the women who lived at a high-supervision residential reentry program, case managers could deny permission at any time to leave for an interview. Case managers’ discretion meant that women had little control over their schedules and were often denied access to a telephone to indicate the change of plans. As Amber, an interviewee in her 30s, explained, “We’re stuck all day [at the reentry facility] unless we get passes to go out. You got to get permission.” Reflecting on the 90-minute pass she was granted at the time of the interview, Amber said, “I think that’s a little harsh.” She went on to compare this lack of autonomy over her schedule to her time behind bars: “I’m doing okay. There’s no other way to handle it. I’ve been locked up.” Women’s autonomy over their own schedules was limited by residential reentry facilities’ oversight, operating within the ever-widening net of carceral control (Prior 2020).

Finding an appropriate interview location was another key challenge. The reentry program directors I spoke to made it clear that their facilities were already at capacity. It would not be possible to conduct our interviews on-site. Furthermore, we wanted to interview women who participated in nonresidential reentry programs to learn about hardships related to housing. We asked prospective interviewees where they would like to meet, choosing locations most convenient to them. Sometimes, this was a public place like a library; other times, a privatized “third space.” Gwen, for one, chose to meet at a McDonalds located on a major artery that ran through the city, easily accessible by public transportation. Christine scheduled her interview at a St. Louis Bread Company (Panera), which her probation officer agreed to drive her to. On several occasions, prospective interviewees were unable to secure transportation and had to cancel their meetings. A Midwestern city with low walkability and limited public transportation meant that lacking access to a car was a form of structural precarity.

Another key feature of structural precarity among formerly incarcerated women related to strained romantic relationships. Intimate relationships with men that operated as pathways to incarceration (Richie 1996; Salisbury and Van Voorhis 2009) sometimes persisted postrelease as roadblocks to reintegration. An excerpt from Amanda’s field notes during an interview with Jackie illuminates one such instance. One December morning, Amanda sat in the lobby of a health services organization across from Jackie. Jackie began the interview by sharing that she has been incarcerated eight times and was unhoused only a few months prior. It soon became clear how the everyday vulnerabilities faced by Jackie’s boyfriend, Carl, shaped her experience of reentry and indeed impacted the interview itself. Less than five minutes after the interview began, Carl barged into the lobby of the health services organization looking distressed. Quoting from Amanda’s abridged field notes:

Carl explains that he had gotten into some kind of conflict with another gentleman outside, who “pulled a blade” on him, and he believes in the chaos, he must have dropped his phone. He tells Jackie that it wasn’t just his phone but attached were three bus passes, his social security card, and other important items…. Jackie initially remains quiet, seemingly unsure of what to say or how to help. She asks him if he left it at the “food pantry” and he responds argumentatively that he couldn’t have lost it there. He continues searching his backpack as Jackie offers additional locations the phone could be, which makes him even more agitated.

…Staff [of the organization] are not intervening at this point…. Jackie’s face is red, and she seems out of breath. Her boyfriend is now shouting at staff, Jackie, and other clients. Jackie is quietly explaining (justifying?) to me…[that Carl has] an addiction to heroin and no family of his own. She tells me that this behavior happens frequently.

The elements of structural precarity that made this population necessary to study were the very factors that made access difficult. Cell phone access was unstable. Substance use posed a challenge. Reentry programs restricted autonomy and made schedules inflexible. Romantic relationships fostered vulnerability. Crucially, I view the onus as on the researcher to navigate these barriers to access. Humility is essential: academic interviews are not—and should not be—a top priority for those in vulnerable circumstances. Early on, we had not properly adapted our strategies to the realities of everyday life for women reentering from jail and prison. The academic calendar did not line up with the lived experiences of the women we wanted to interview. When we were in class, we missed calls. We left town for holiday breaks. We underestimated the challenges of public transportation. The privileges afforded in the ivory tower created blinders to the practicalities of daily life among a vulnerable population.

It became increasingly clear that interviews conducted on-site at a reentry organization would be the most effective approach. Drawing once more on strategies of institutional legitimacy, I connected with a leading staff member of a reentry organization via e-mail. After a few phone calls, we were granted permission to attend weekly meetings to recruit for our interviews. When I observed one such meeting, I was struck by how many women were interested in participating in the study. Interviews would be conducted then and there. No scheduling was required. This final recruitment coup offered us the most fruitful lesson of all: we needed to show up. In populations facing structural precarity, physical presence is the surest means to access (e.g., Duck 2015; Stuart 2016). By the end of the study, we received more snowball referrals than we could manage.

This may sound like an endorsement for ethnography. After all, showing up is part and parcel of ethnographic fieldwork, and in doing so, we circumvented a few of the previous barriers to access. Observational fieldwork may generate rapport and participants’ buy-in where flyers and referrals may not. This is especially important for research across social boundaries. Nevertheless, an interview study with a structurally precarious population does not require an ethnographic component. For one, reentry is diffuse. There is no unified social world of reentry in the same way there is inside prison walls. Indeed, scholars have noted the relative stability of incarcerated populations compared to their formerly incarcerated counterparts (Comfort and Grinstead 2004; Sufrin 2017) insofar as access to housing and medical care. The experience of reentry permeates across communities—woven into public, private, and penal spaces—de facto limiting the contributions of a single-sited or even multisited ethnography. Furthermore, an ethnography of a reentry program might exclude the most vulnerable individuals, particularly those without housing, transportation, social safety nets, and support for substance abuse. Ideally, whether using interviews or observational methods, successfully accessing a structurally precarious population as an outsider involves a combination of institution-based recruitment alongside recruitment of those unaffiliated with social service organizations.