Studying Homeless and Incarcerated Persons: A Comparative Account of Doing Field Research With Hard-to-Reach Populations

Abstract: In this article, I discuss the process of conducting research with two vulnerable and hard-to-reach populations (homeless and incarcerated men) in three research locations characterized by varying levels of gatekeeping: a prison, public streets in an urban city, and a residential facility for homeless men. I argue that, despite the obstacles to independent research that gatekeepers (officials who can grant or deny researchers access to participants) pose, research with vulnerable, hard-to-reach populations in different field sites reveals some of the benefits of using field sites characterized by gatekeeping and strict rules to which researchers must adhere. Many of these benefits, however, go unacknowledged in discussions of access in qualitative studies—especially in the penological literature. I conclude that, instead of shying away from qualitative prison studies, researchers should take advantage of the benefits that prisons offer as field sites.

Key words: gatekeeping; field research; interviewing; grounded theory methodology; access; prison; prison research; homelessness

Scholars have devoted a significant amount of attention to the difficulties in accessing vulnerable populations for research purposes (ANDERSON & CALHOUN, 1992; PAWELZ, 2018; SCHLOSSER, 2008; UMAMAHESWAR, 2014). Even once access is gained, studying vulnerable populations can be particularly difficult because members of these populations often have low literacy levels (which complicates the typical process of obtaining written informed consent), and their ability and/or willingness to participate in research is often contingent on factors out of the researchers' control. For instance, in the prison setting, certain participants (for example, those housed in administrative segregation) may be preemptively prohibited from participation by prison officials. Another example can be found in the work of researchers attempting to study homeless populations: The instability that often characterizes the lives of those experiencing homelessness means that researchers cannot reasonably count on participants consistently being able and willing to participate in a research study (KIDDEY & SCHOFIELD, 2011). [1]

Drawing on an ongoing sociological study in which I seek to study the tie between incarceration and homelessness, I present a reflexive account of the process of conducting qualitative research on two vulnerable, hard-to-reach populations: homeless and incarcerated men. My broad goal in the study is to investigate these men's constructions of masculinity in the context of further understanding why so many homeless men end up incarcerated and vice versa. More specifically, using in-depth, semi-structured interviews, I explore whether the men's attempts (and perhaps failure) to conform to conventional masculine ideologies—could explain the "nexus" (GOWAN, 2002) of homelessness and incarceration. In this particular article, I examine how conducting research with vulnerable populations in starkly different field sites sheds light on issues that previous research has overlooked related to gatekeeping (ANDERSON & CALHOUN, 1992; MILNE, 2005; PAWELZ, 2018; SCHLOSSER, 2008; UMAMAHESWAR, 2014) as well as the ethics (UMAMAHESWAR, 2014) and risks (PAWELZ, 2018; SCHLOSSER, 2008) of conducting fieldwork with vulnerable and hard-to-reach populations. [2]

ANDERSON and CALHOUN (1992) believe that the absence of gatekeepers (people who have the power to grant or deny scholars access to research participants) on the street (understood as the public domain in cities, such as sidewalks, parks, and other open spaces) renders studying homeless populations in this context easier than studying them in shelters or other sites that are characterized by stricter gatekeeping. On the other end of the spectrum, prison scholars across the world (BOSWORTH, CAMPBELL, DEMBY, FERRANTI & SANTOS, 2005; SCHLOSSER, 2008) have highlighted the difficulties in conducting field research in a prison setting because of the strict gatekeeping that characterizes most detention facilities. Even outside the prison setting, some researchers have gone even further to argue that institutional reviews of research (a form of gatekeeping) are shaped less by considerations of the ethical treatment of participants and more about power and control (ROTH, 2005). All these views reflect an underlying assumption that conducting research in settings that lack gatekeepers is easier. In this article, however, I argue that there are benefits to gatekeeping of which scholars interested in studying vulnerable and/or hard-to-reach populations should be aware. [3]

My purpose is to highlight the ethical and methodological issues that researchers contemplating field research with multiple vulnerable populations, and using multiple sites, should expect to encounter. I argue that, while research in sites such as prisons (which require approval from gatekeepers) is daunting in many ways, studying vulnerable populations in settings that lack gatekeepers (such as public, open spaces) is not necessarily easier or more useful. I conclude that researchers interested in pursuing qualitative field work with hard-to-reach populations—particularly in prisons—should be aware of the difficulties such research poses, but that they should also seek to exploit some of the advantages that prisons offer as fieldwork sites. Such research is particularly important at a time when researchers are overwhelmed by the difficulty of conducting field research with prison populations (SCHLOSSER, 2008; UMAMAHESWAR, 2014; WALDRAM, 2009). [4]

Although scholars have discussed issues related to accessing field sites generally (BROADHEAD & RIST, 1976; DUKE, 2002; REEVES, 2010), many of these researchers have focused on only small segments of vulnerable populations, perhaps because of the challenges of accessing and studying multiple hard-to-reach groups in the field. Instead of reviewing the large amount of research on access in fieldwork, I limit my review in this section to literature discussing the qualitative research process involved in studying the two groups relevant to this study: Incarcerated and homeless populations. [5]

After a period during the 1950s and 1960s that scholars (SIMON, 2000, p.285) labeled the "golden age of prison sociology” in the United States, there has been a sharp decline in qualitative field research that uses prisons as field sites. Institutional review boards (IRBs) are cautious about approving research with prisoners—a population that many boards see as among the most vulnerable (BOSWORTH et al., 2005)—and prison populations in the past few decades have become increasingly difficult for researchers to study (ibid.). Scholars have expressed disappointment in the shortage of rich qualitative prison research (KREAGER & KRUTTSCHNITT, 2018), but many simultaneously recognize that the decrease in such research is partially explained by the uphill battle scholars face when seeking approval for access from state agencies as well as educational institutions (BOSWORTH et al., 2005; UMAMAHESWAR, 2014; WALDRAM, 2009). [6]

Much of the methodological literature (SCHLOSSER, 2008; UMAMAHESWAR, 2014; WALDRAM, 2009) on qualitative prison work has thus focused on how access to prisoners requires a multi-step, arduous and incredibly time-consuming process that requires approval from researchers' IRBs, and prison administrators before such research can even begin. Additionally, research in U.S. prisons typically requires approval from the Department of Corrections—the governmental body in charge of overseeing and managing incarceration. Thereafter, maintaining access has proven difficult as well: Logistically, threats of lock downs that suddenly (and at least temporarily) terminate access loom large, prison routines are constantly changing, and prisons face high rates of both prisoner and staff turnover, resulting in an environment characterized by constant flux. Despite the ever-changing nature of the prison environment, the prison regimen must always be honored. This means that researchers may be permitted to interview only at certain times and for certain durations to accommodate staff shift changes and prisoner movements. During any given field trip to a prison, participants may also be made to wait for long periods before being interviewed because they are permitted to move only at designated times. [7]

These logistical constraints are tied to a more fundamental problem related to maintaining access to participants: Prisoners' trust must be earned for qualitative prison research to be successful (BOSWORTH et al., 2005; SCHLOSSER, 2008; UMAMAHESWAR, 2014; WALDRAM, 2009). When participants are brought into an interview room in a prison (often after waiting for an extended period without being told why they are there or with whom they are meeting), the burden falls on the researcher to establish a relationship of cooperation and trust with the participant. Failing this, the participants may simply choose to leave the interview, as they have every right to do. Further, prisons in the U.S. are typically "total institutions" (GOFFMAN, 1961, p.4): They are same-sex institutions whose residents are both physically and socially segregated from the outside world. A result of this segregation is that the prison becomes a self-contained world in which word travels quickly. If scholars do not build trusting relationships with participants early on in the research process, they risk losing access not only to those prisoners, but also to future participants who hear from others that they should steer clear of a research project. [8]

Scholars studying incarceration have expressed dismay about the decline in qualitative prison research (KREAGER & KRUTTSCHNITT, 2018; WACQUANT, 2002), but they have also called for researchers to make a concerted effort to revive the rich prison ethnographies of earlier decades that laid the groundwork for much of our knowledge about prison life. In light of a growing interest in reviving the qualitative prison research that thrived many decades ago (GIALLOMBARDO, 1966; HEFFERNAN, 1972, WARD & KASSEBAUM, 1965), researchers (BOSWORTH et al., 2005; SCHLOSSER, 2008) have actively worked toward the development of strategies that facilitate qualitative prison research. Given the difficulties in reaching incarcerated populations, for instance, some researchers (BOSWORTH et al., 2005; UMAMAHESWAR, 2014) have promoted the use of written correspondence as a way to study incarcerated populations from afar. In the hopes of guiding future prison work, other researchers (SCHLOSSER, 2008) have discussed some of the unique features of conducting qualitative research in a high-risk research site such as a prison. [9]

Underlying all of this work, however, is the view that strict gatekeeping by state agencies and educational IRBs has rendered prison research incredibly difficult. My argument in this article is not intended to be a refutation of this point; rather, I argue that, despite the immense benefit of existing methodological discussions on qualitative prison research, the current literature has overlooked the (perhaps unintended) positive consequences of this level of rigidity in the research process. By highlighting certain features of prison research that have often been ignored, and by comparing the process of conducting prison research with conducting fieldwork with vulnerable populations in other sites, I discuss how prison research may actually be easier than research in settings that have little to no gatekeeping. [10]

Like prisoners, homeless persons constitute a vulnerable population. Persons experiencing homelessness are among the most seriously disadvantaged members of American society, often suffering from acute mental and physical health issues, severe poverty (LIU, STINSON, HERNANDEZ, SHEPARD & HAAG, 2009), and difficult histories of drug and alcohol abuse (VAN GEEST & JOHNSON, 2002) as well as histories of criminal behavior (STEIN & GELBERG, 1995). [11]

Many scholars studying homeless populations have used shelters as fieldwork sites (LIU et al., 2009). A single site is often helpful in structuring fieldwork that would otherwise be unpredictable, but using shelters as research sites often requires access negotiations with officials at social service agencies (TAYLOR, 1993) as well as staff members at the shelters themselves. Persons living in shelters also represent a distinctive sub-group of the homeless population insofar as they are both willing and able to seek the assistance of the shelter system. While rendering the process of conducting research with homeless populations easier in some ways, studying only homeless persons residing in shelters limits our knowledge about the true scope and experience of homelessness. [12]

Other scholars (ANDERSON & CALHOUN, 1992; PASSARO, 1996; SNOW & ANDERSON, 1993) have taken directly to the streets to study homeless populations. ANDERSON and CALHOUN (1992) argued that although few researchers choose simply to approach homeless persons on the street, this approach can eliminate concerns about access when conducting qualitative fieldwork. Most obviously, because homeless persons can simply be approached on the street, the researcher need not worry about the lengthy process of seeking access to a research site, only to be ultimately denied such access by gatekeepers at the site (ibid.). Studying persons experiencing homelessness in public spaces thus represents fieldwork that is in some ways directly opposed to the nature of fieldwork in prisons: It relieves researchers of the burden of negotiating access with gatekeepers who are deeply protective of their institutions. It thereby permits researchers a degree of independence that is very difficult to find when conducting research in closed institutions such as prisons. [13]

KIDDEY and SCHOFIELD (2011), however, note that studying persons experiencing homelessness in this way requires a commitment to remaining flexible during the research process. They argue that researchers cannot reasonably and consistently rely on persons experiencing homelessness to be able and willing to participate in a study. Specifically, because their circumstances are unpredictable and ever-changing, their willingness and/or ability to participate in a study one day is not necessarily predictive of willingness and/or ability to participate the next (ibid.). [14]

When studying homeless populations, interview location is also an important consideration (ECKER, 2017) for ethical reasons. TORO (2006), for instance, noted that interviewing subjects in public places raises concerns about confidentiality and safety. While some researchers have thus flagged the difficulties in studying homeless populations on the street, methodological research on access in particular has continued to point primarily to the difficulties that gatekeepers pose to researchers attempting to study hard-to-reach populations rather than also discussing the benefits that gatekeeping can offer researchers. [15]

As I describe in the next section, the current study uses three separate data collection sites, allowing for a comparison of how the qualitative research process differs across settings that vary widely in the level of gatekeeping and flexibility they present. After describing the study on which this article is based, I point to the ways in which current research has failed to acknowledge some of the benefits of gatekeeping, especially in the context of prison research, focusing instead almost exclusively on the barriers that gatekeeping poses. [16]

Scholars have focused extensively on the incarceration boom of the last few decades, highlighting the causes for the immense spike in American prison populations (BLUMSTEIN & BECK, 1999) as well as the consequences of mass incarceration (HAGAN & DINOVITZER, 1999; WESTERN & WILDEMAN, 2009). Very recently, however, the United States has entered a new phase of decarceration (PHELPS & PAGER, 2016), as politicians and the public alike recognize the perils of sending massive numbers of Americans to prison. At this juncture in American penal history, it is crucial that we pay even greater attention to the reentry issues that prisoners face upon their release. One such issue is the risk of homelessness: Prior research (GOWAN, 2002) has indicated that there is a nexus of homelessness and incarceration, such that those who have spent time in prison face a higher risk of housing insecurity, while those that have experienced homelessness are more likely than others to end up in prison (METRAUX, ROMAN & CHO, 2008). [17]

A separate body of research on men and masculinity has highlighted how men's crimes reflect their conformity to prevailing discourses regarding masculinity (NEWBURN & STANKO, 1994), and related research has also documented extensively the "hyper-masculine" nature of the men's prison subculture (SABO, KUPERS & LONDON, 2001). Despite the overlap in the experiences of homeless and incarcerated men, however, there has been no qualitative research comparing them. Combining research on the incarceration-homelessness link with literature on prisons and masculinity, I aim to fill this gap by comparing homeless and incarcerated men's definitions of masculinity. More specifically, using semi-structured, in-depth interviews, I investigate the extent to which men's pathways into and out of homelessness and prison reflect their conformity to, or departure from, dominant ideologies regarding masculinity and adulthood. I decided to use semi-structured, in-depth interviews to facilitate a level of flexibility that allowed participants to shape the direction of the interview while remaining on-topic and ensuring that the data I collected across interviews were comparable (RABIONET, 2011). The specific research questions I investigate are:

I use the grounded theory approach in data collection and analysis in this study. Per the tenets of this approach, data collection and analysis is merged, such that I use data that I have already collected and analyzed to revise and update future data collection efforts. My use of this approach also requires that my fieldwork is flexible and open to changes in the research process that are shaped by early data collection and analysis. Indeed, my use of a semi-structured and open-ended interview guide instead of more rigid data collection instruments (such as a survey or structured, close-ended interview schedule) reflects my own willingness to let participants shape the direction of the interview. This, I believe, is particularly important in a study focused on the participants' constructions of their own identities. [19]

My analysis of the data is based on the inductive search for patterns of similarities and differences in the data, and my ultimate goal is to produce a synthesized, theoretical account of the data. While every version of the grounded theory approach shares certain common principles, I use CHARMAZ's (2003) version of the grounded theory approach to data collection and analysis in particular. In this version, data obtained using the grounded theory method reflect both the researcher's and the participant's mutual construction, and the researcher is affected by her interaction in her participant's world. Given the exploratory nature of this study, as well as its focus on how participants conceptualize and enact their own identities as adult men, the emphasis on identity construction and symbolic interactionism in CHARMAZ's (2003) version of grounded theory method is especially appropriate. [20]

Although data collection for this project is still ongoing, this article is based on 31 interviews already completed in 3 different sites: 4 interviews with men living in a residential facility in an urban city in Northeast U.S., 3 interviews with homeless men on public streets in the same city, and 24 interviews at a state prison, again in Northeast U.S. I describe each of these sites in turn. [21]

The Eagle Foundation (a pseudonym chosen to protect the confidentiality of the participants) is a non-profit organization located in a major urban city near the East Coast of the U.S. The goal of the organization is to assist men who have been homeless and/or incarcerated in their efforts to reintegrate into mainstream society. To this end, the organization serves as a residential facility in which the men are given shelter and food as well as the hub from which they are exposed to a wide range of job training programs and job opportunities. Most men at the Eagle Foundation have experienced both homelessness and incarceration. They are enrolled in a program that typically lasts 9 months to 1 year, and they are provided housing and food during this period. [22]

Before initiating any contact with the Eagle Foundation, I obtained IRB approval from my institution for this research. I then began my relationship with the Eagle Foundation as a volunteer literacy tutor for men who were preparing to take a food handlers' exam offered by the city in which the foundation is located. Following this, I sought and received approval from the facility to recruit residents for participation in my research. Specifically, I requested permission from staff to interview men residing at the facility, which I was granted as long as I made it clear to the residents that their participation in the study was voluntary and that I would maintain their confidentiality. I explained that I would protect participants' confidentiality in the informed consent form, and I also stipulated in the form that they could withdraw their participation at any time. Finally, the form indicated that participants would not receive any direct benefits from participation, nor would participation affect their eligibility for parole or any other program. [23]

For reasons that I discuss later, despite my preference for a purposive sampling strategy based on my commitment to the grounded theory approach, I employed a combination of convenience and snowball sampling at the Eagle Foundation. With the approval of the facility, I recruited participants for the study by requesting to speak with them about my project during lunch hour at the facility. I mingled with the residents of the facility as they lined up for lunch and I provided them with letters of introduction that provided more detail about the project. These letters also contained a tear-off response that they could return either to me or to staff officials indicating their interest in participating in the study. I then arranged a time with staff officials during which I could interview interested residents in private rooms at the facility. After giving the participants more detail on the project's scope and goals, I orally communicated the content of the informed consent form. Before beginning the interview, I requested participants' signatures on two copies of the informed consent form, one of which they kept and one of which I retained for my own records. [24]

In addition to studying men at the Eagle Foundation, I decided that it was important that I also study men who lack the stability and structure that the facility offers its residents. I was specifically concerned about the selection bias that could result from studying only those homeless men who are able and willing to seek the assistance of an organization such as the Eagle Foundation, which offers residents food and lodging. To ameliorate the risk of this selection bias, in addition to interviewing men living at the Eagle Foundation, I decided also to interview homeless men living on the streets located in the same city as the Eagle Foundation. I pursued this latter strategy based on previous scholars' (ANDERSON & CALHOUN, 1992) belief that it is both practical and effective—a conclusion that I discuss shortly. [25]

Using a convenience sampling strategy, I approached men who were panhandling and/or those who had signs indicating that they were homeless. I chose to approach only these men because relying on other more implicit indicators of homelessness (e.g., mode of dress) can be inappropriate and give rise to problematic class-based biases. Before beginning the interview, I requested the opportunity to speak to them about my project. I orally conveyed to them the information that I had on the informed consent form that I used in each of my field sites, and I asked them explicitly if they were willing to participate in the research. If they agreed, I asked them if they were comfortable being recorded before I continued with the interview. Thus far, the men I have interviewed have been open and willing to speak with me, although (as I will discuss later) the data from each interview have varied in quality. [26]

Finally, this study draws on data obtained from incarcerated men in a state prison also located in the Northeastern part of the U.S. This prison houses a diverse range of offenders, including serious, violent offenders who are housed in the closed-security unit and minimum-security offenders who are offered community service. The prison also houses inmates with addictive behaviors in a therapeutic community. I did not exclude any group of offenders from participation in this study except those men who were housed in administrative segregation. The exclusion of prisoners housed in administrative segregation was the only limitation that the prison imposed on me with regard to which prisoners I could interview. Participants were convicted of crimes that were both violent (e.g., rape and first-degree murder) and non-violent (e.g., drug offenses). [27]

I was given access to prison records containing the names of every man incarcerated at the facility along with basic demographic information (race and age) as well as offense information (data of booking, date of admission, date of earliest possible parole, offense description, and maximum possible sentence). Using these records, I pursued a purposive sampling strategy based on attaining diversity on theoretically important variables (race, age, offense committed, and sentence length). Although I did not analyze the records themselves as part of the study, they were invaluable because they permitted me to sample purposively, and they also served as a basic validation check for elements of the participants' narratives, such as information on their conviction and sentence length. [28]

Initially, I pursued a sampling strategy whereby I mailed potential participants a letter introducing myself and the project. I asked the prisoners to tear off part of the letter indicating their interest in participating and return it either to me by mail or to a prison staff member. This strategy, however, proved to be largely unsuccessful, with very few men responding to the letter. The low response rate was unsurprising because the process required that the men either spend money on their own envelope and stamp (both of which can be very costly in the prison setting) or return the slip to a prison staff member that they may not know and/or trust. While I would have much preferred to send the participants a pre-stamped envelope in which they could return their slip to me, I was not permitted to do so by the prison. I was thus forced to change my sampling strategy because of the low response rate and the high likelihood of a problematic selection bias. Specifically, I was concerned that only inmates who were eager enough to participate in the study that they were willing to trust a prison staff member or pay for the stamp and envelope were participating. I thus amended my sampling strategy so that I liaised with a prison staff member to set up meetings with potential participants without first sending them a letter. [29]

After setting up these initial meetings with the prison staff, I was usually assigned a small, private room in the part of the prison in which prisoners meet with psychologists, parole officers, and other professionals for official business. In some instances, I was escorted to a different wing in the prison, but I was still granted a private space for each interview. For safety purposes, correctional officers were posted outside the room, but to preserve confidentiality, none were present in the room itself. While I communicated to participants that only I would have access to their narratives, it is important to note that I had no choice but to disclose to staff members the names of the prisoners I wished to interview because prison rules require that the location and movement of each prisoner is always accounted for. In the initial meeting, I introduced myself and the project and asked the men if they were willing to participate in the study. If they were comfortable with participating, I proceeded with obtaining signed informed consent process before beginning the interview. Only one prisoner refused to meet with me completely and only one prisoner with whom I met refused to participate in the study. [30]

In all three settings, the interview consisted of questions regarding the participants' thoughts on what it means to be a man, their recollections of their early life experiences and their transition to adulthood (with an emphasis on those that shaped their views on masculinity) and their prison and/or homelessness experiences. I did not dwell on any of the men's criminal histories except to the extent that they were tied to their definitions or enactments of masculinity, but I did ask each participant in the prison about the circumstances that resulted in their incarceration. [31]

Before proceeding to a discussion of the benefits of gatekeeping, it should first be noted that my experience affirmed ANDERSON and CALHOUN's (1992) view that studying individuals on the street is simpler in one sense than attempting research in an organizational setting. Access to the prison in this study took months to arrange, and access to the Eagle Foundation was a similarly long process that required a great deal of negotiation and compromise with staff at the facility who did not have formal guidelines to which to refer when determining how to evaluate my research proposal. [32]

My experience with the Eagle Foundation sheds light on the process of conducting research in a setting characterized by a level of gatekeeping in between that of a prison and that of the street. In many ways, the Eagle Foundation proved to be the most difficult setting in which to conduct research. The facility officials were responsible for approving the research, and I relied on the staff at the facility to arrange access to the residents and interview rooms on the day of interview. Yet because the organization had a less formalized procedure for handling external research requests than a prison typically does, the research process was more unpredictable than prison research, despite coming with many of the same gatekeeping obstacles. [33]

For example, much like at the prison, officials at the shelter were particular about when/where I could collect my data and they also had sole authority to grant or deny me access to resident records. In the case of the Department of Corrections, however, there were formal policies that structured how data access requests were handled. At the Eagle Foundation, there was no formal data access request policy, and my informal request for access to some resident information was denied because of concerns about protecting residents' confidentiality. Although I was able to obtain rich narratives from the men living at the facility because I was granted a private, secure room in which to conduct the interviews, the absence of formal policies made seeking and maintaining access to the site much more stressful than was the case at the prison. [34]

Further, although I was sensitive to the confidentiality reasons behind the denial of access to residents' records, the gatekeeping in this setting severely curtailed the independence I could exert on my fieldwork. Without records to use to target specific participants using a purposive sampling strategy, I had little choice but to pursue a combination of snowball and convenience sampling by recruiting residents I met in the facility and asking them to let their fellow residents know of the study. This level of negotiation and compromise was necessary to gain access to the facility for research purposes at all. [35]

In contrast to the negotiation required to gain access to the Eagle Foundation and the prison, ANDERSON and CALHOUN (1992) have argued that collecting data from men living on the street requires little more than approaching the men and striking up a conversation. Indeed, as mentioned earlier, this was generally the method I used to interview homeless men on the street: I simply approached men sitting on street corners who were panhandling and/or who had signs indicating that they were experiencing homelessness, and I asked them if I could speak with them about my project. This ease of initial access, however, masks the very real difficulties involved in conducting fieldwork on the street—many of which involve the ethical decisions that researchers conducting fieldwork of this nature should be prepared to encounter. Although it was fairly easy to approach men on the street, doing so introduced a number of problematic elements that merit more discussion than they have received in the existing research. Next, I describe some of the more complex issues that prior researchers have failed to examine when discussing the ease of conducting research in the absence of gatekeeping (and conversely, the difficulties of conducting research in the presence of gatekeepers). In particular, I discuss four key areas: 1. ethical issues, 2. researcher vulnerability and research ethics, 3. physical risks and researcher safety, and 4. quality of the data obtained. [36]

A central issue in discussions of research with vulnerable populations is that of consent (SCHLOSSER, 2008; UMAMAHESWAR, 2014). As mentioned, prior research has indicated that both the populations in this study may have low literacy levels, histories of drug and/or alcohol abuse, mental and physical illnesses, and high rates of poverty. It was therefore very important that I handled the process of informed consent thoughtfully in each of the field sites used in this study. In this section, I discuss issues of informed consent that emerged in the different settings. [37]

Obtaining informed consent from residents at the Eagle Foundation was fairly straightforward. I first submitted the informed consent form that my IRB had approved to the facility staff. With the permission of the staff, I orally communicated the content of the form to the participants before requesting their signature on the form and inviting them to keep a copy of the form for their own records. The informed consent process at the facility was thus simple and predictable. [38]

In contrast, obtaining informed consent from persons living on the street was much more complicated. In addition to the more obvious issue of managing participant confidentiality in open spaces, handling informed consent when conducting research on the street—a quintessentially unsupervised research setting—was very difficult, in part because field research of this nature was very unpredictable. With the approval of my institution's IRB, I initially planned to explain orally the content of the informed consent form before requesting a participant's signature (as I did in the other two sites). It quickly became clear, however, that this plan was infeasible. In one conversation with a participant on the street, as soon as I explained my role as a researcher, he began recounting his life story. The spontaneous and unstructured nature of our interaction disrupted the typical step-by-step procedure of seeking informed consent. Additionally, homeless men living on the streets live at the periphery of society, and are often in conflict with other homeless men, passersby, and (perhaps most problematically) the police. Requesting a formal signature thus also entailed a more burdensome request than doing so in other fieldwork settings because participants in conflict with the legal system and/or other state agencies may be wary of the implications of putting their name on an official institutional form. [39]

After considering all these issues, I decided that I would administer only oral informed consent—an amendment that required further IRB approval. I submitted a proposed revision to the IRB, explaining the difficulties of obtaining written informed consent, and I asked to switch to oral consent, whereby I would explain the content of the form and gain participants' permission to proceed with the interview orally instead of through a written signature. I resumed data collection once I had received approval of the change to oral consent. Although administering informed consent orally did not fully overcome some of the problems I have highlighted, it was a safer, more practical alternative to signed informed consent, and one that was more sensitive to the unique circumstances of men living on the street. [40]

Studying prisoners comes with its own set of issues that researchers must confront when administering informed consent. Researchers must, for instance, address the possibility of low literacy levels, ensure that participation is genuinely voluntary, and be prepared for participants to make requests of them because of their status as privileged "outsiders" (SCHLOSSER, 2008; UMAMAHESWAR, 2014)¹⁾. Decisions about how to handle these issues are typically made at the beginning of the research process, however, such that once negotiations with the prison and IRB are complete, the informed consent process is typically predictable and consistent. [41]

In the current study, for example, the informed consent form contained language that both my IRB and the prison's review board had approved that covered many of the issues described earlier. The form was written at the eighth-grade reading level (as was the form I used in the other sites), it stipulated that participants would not benefit directly from participation, and it assured participants that they could withdraw their participation whenever they wished. The form also included stipulations that participation or non-participation would not affect eligibility for parole or any other government-run program (such as welfare). At the time of the interview, I followed an informed consent process very similar to the one I used at the Eagle Foundation: I met with my participants, introduced myself and my project to them, asked them if they would be willing to answer a few questions, and orally communicated the contents of the informed consent form. I then obtained written informed consent and provide participants with two copies of the informed consent form (one for them and one for myself) to sign before beginning the interview. [42]

While recognizing the problematic elements of administering informed consent to captive, often low-literacy participants (SCHLOSSER, 2008; UMAMAHESWAR, 2014), it is thus equally important to acknowledge that the absence of gatekeepers does not necessarily yield an easier (or less problematic) informed consent process, as my research on the street indicated. [43]

JEWKES argued that prison researchers are often problematically "silent on their own agency, identity management, and survival and socialization strategies" (2014, p.387) despite focusing extensively on these features of their participants' lives. She thus called for prison researchers to be more transparent in disclosing the many emotions they experience when conducting research in prisons. Particularly relevant for the purposes of conducting qualitative research with vulnerable populations is a discussion of ethics and researcher vulnerability. [44]

Researchers studying both incarcerated and homeless populations should expect to confront ethical questions in studying both populations that are difficult to resolve and that give rise to complex emotional responses. Members of both these groups may, for instance, see a researcher as a privileged outsider who is in a position to help them financially, professionally, and/or emotionally. However, prisons are highly formalized and regimented institutions, and the bureaucratic nature of prisons can be very helpful in addressing these requests when they do arise. In many cases, the researcher is often able to refer to and cite prison policies that usually prohibit researchers from offering participants assistance outside the scope of the research project. In the unsupervised setting of the street, the researcher must address such requests entirely independently. In the current study, for example, one participant requested money after the interview. When I explained that I was unable to provide monetary compensation, he instead asked for a cup of coffee, a request that I honored. [45]

Researchers planning on studying homeless populations in this setting should be prepared to confront feelings of empathy, sympathy, and even distress when faced with requests of this nature, all of which are made more acute because of the absence of formal rules to rely on when constructing an appropriate response. In studying homeless men on the street, I was unable to formulate a principled, consistent rule for handling requests of this nature because the experience of interviewing in public spaces was unavoidably unpredictable and inconsistent. In contrast, I was able to draw a sense of security in the rules of the prison that gave me a formal, consistent way to handle requests from participants for my help. Despite the harrowing emotional effects of conducting field research in prisons (LIEBLING, 1999), scholars should also be aware that the bureaucratic structure of prisons can be helpful in combating the feelings of distress and frustration that arise when participants ask for assistance. [46]

Persons experiencing homelessness also frequently have high rates of drug and/or alcohol use (LIU et al., 2010). In this study, one participant on the street freely admitted that he was under the influence of heroin at the time of the interview. Researchers who study homelessness must confront the ethical and practical question of whether or not to exclude participants who are obviously under the influence of alcohol and/or drugs. In this study, my decision not to exclude such participants was based on a number of factors. First, and perhaps most importantly, I sought out homeless persons living on the street to avoid a selection bias in my sample, and I believe that this goal required excluding the fewest categories of homeless persons as possible. As such, until and unless I was concerned for my own safety or about my ability to communicate with the participant, I did not exclude anybody living on the street from participating in the study. [47]

Secondly, as KIDDEY and SCHOFIELD (2011) have argued, there are ethical as well as methodological considerations involved in excluding homeless people who are under the influence of drugs and/or alcohol. I believe that qualitative research should be as inclusive as possible—and in fact, as was the case in KIDDEY and SCHOFIELD's research, drug use and addiction proved to be an important feature of my participants' narratives. Finally, it is important for researchers studying any vulnerable population to be aware that addiction and drug use does not always manifest overtly and, methodologically speaking, there may be no definitive way to exclude participants under the influence of drugs and/or alcohol. Indeed, many of the participants I spoke with at both the Eagle Foundation and the prison noted that drugs were freely available in those sites as well. Asking a participant to disclose whether or not he/she is under the influence of drugs and/or alcohol is both unreliable and arguably unethical to the extent that it requires disclosure about criminal behavior that the participant may not be comfortable sharing. Trying to glean from the participant's behavior whether he/she is under the influence of drugs and/or alcohol is equally (if not more) unreliable. As such, for both ethical and methodological reasons, I concluded that it was important not to exclude participants who were drunk and/or under the influence of drugs, even if this resulted in difficulties in validating their stories. [48]

Of course, there are countervailing ethical considerations that may tilt researchers toward excluding participants under the influence of drugs and/or alcohol. Perhaps the most salient of these include the need for researchers to protect themselves. Interviewing men in public spaces who are under the influence of drugs and/or alcohol poses an obvious risk to the safety of researchers. Partly to protect my own safety (a feature of the study that I discuss later), I chose not to interview homeless men who had congregated in groups on the street. Further, interviewing the men in groups would not only be methodologically problematic because of how the presence of others may affect their narratives: doing so would also have posed ethical concerns related to confidentiality and anonymity. [49]

There is no doubt, however, that the interviews conducted with homeless persons on the street had less ethical integrity than those conducted in more formal settings (ECKER, 2017). As ECKER has noted, interviewing homeless persons in public spaces makes it difficult to maintain their confidentiality. I tried to maximize the ethical integrity of this study by offering participants in all three settings the same protections, which included (among others) a guarantee not to make their identities known in any research product, as well as an explicit explanation of their rights to withdraw their participation at any time and to skip any question that they did not wish to answer. It can be argued, of course, that men under the influence of alcohol and/or drugs were unable to consent to participation in such a study. As I mentioned earlier, however, it was very difficult to determine in advance whether any given participant was under the influence of drugs and/or alcohol, and after careful reflection, I ultimately agreed with other researchers (KIDDEY & SCHOFIELD, 2011)) who have argued that it is important to include the voices of such participants in the body of work on homelessness. [50]

In addition to the emotional vulnerability that researchers face (JEWKES, 2014), the current study underscores the shifting levels of researchers' physical vulnerability based on interview settings, a feature of field research of which scholars appear to be aware. In this section, I describe how the nature of the different field sites had consequences for my own safety as a researcher as well as for the quality of the data I was able to obtain from each site. [51]

ANDERSON and CALHOUN's (1992) conclusion regarding the ease of approaching homeless people on the street glosses over the fact that approaching them in this way can be dangerous and, for that reason, particularly impractical. In one case, the participant with whom I was speaking on the street had an aggressive dog that tried to bite any approaching pedestrian. I was constantly distracted by the dog, and this interview ended with another homeless man joining us, with another aggressive dog. Even setting aside issues of confidentiality in such a situation, the setting appeared risky enough that I ended the interview prematurely and left. In contrast, I did not have any concerns about my safety when conducting interviews at the Eagle Foundation. The facility had security guards (although they were much less visible than the correctional officers in the prison) and conducting the interviews in a private room granted me a level of comfort and focus that was entirely absent in my interviews on the street. [52]

It is important here to note that my experiences with this study are shaped by my own status as a relatively young, female researcher. The significance of this status became evident in my first interview with a homeless man living on the street. A passerby looked at me and then interrupted my conversation with the participant (a 30-year-old, White man) to ask him, "You getting sex?" He then gave the man a thumbs-up sign and left. Women studying men on the street should expect to encounter interactions such as these while in the field, and it is entirely possible that researchers with different demographic traits would not experience some of these challenges. As mentioned earlier, I chose not to interview homeless men who gathered in groups on the street in part to protect my own safety. The point at which researchers believe that risks to their safety outweigh the ethical and methodological benefits of inclusivity will likely vary from one scholar to the next based both on personal histories as well as demographic variables. In my case, approaching a group of homeless men on the street seemed imprudent, largely because I am a relatively young, female researcher. It is also possible, however, that this same status permitted the men to express a level of vulnerability that they would not have been comfortable expressing in front of a male researcher. Homeless men living on the street may also have felt a greater level of safety speaking to a young woman than they would have speaking to another man. [53]

The prison environment can be volatile, and the threat of lock-downs and riots looms large (SCHLOSSER, 2008). To be clear, it is not my intention to understate the extent of these challenges. It is hard not to become frustrated with the strict rules that one must abide by when doing prison research; but comparing this research process with the purportedly easier (ANDERSON & CALHOUN,1992) task of conducting research with populations lacking gatekeepers highlights the benefits—rather than only the challenges—of conducting research in a prison setting. Acknowledging and exploiting the advantages of a highly supervised research setting can assist researchers in overcoming their hesitation in conducting qualitative fieldwork in prisons. The current study sheds light in particular on the extent to which prisons may actually be a safer venue for research on vulnerable populations than prior research has suggested. [54]

The prison environment can be dangerous, but prison officials understand this fact better than most. The safety of "outsiders" such as researchers is therefore a priority for most prison staff, and researchers are afforded a level of protection that is entirely absent when conducting fieldwork in public, open spaces. Once access is gained, the institutional requirements of prison research in many ways make interviewing in a prison setting much more predictable and safer than conducting comparable research with participants on the street. Although interviewing in prisons often requires a great deal of patience from researchers who are expected to wait for long periods to accommodate prison routines, researchers are ultimately often granted a reliably safe space in which to conduct the interviews. In the current study, interviews were scheduled through the prison administrators, who made available a private room for the interviews and arranged for correctional officers to escort prisoners to and from the room. As I discuss next, in addition to minimizing the physical risks the researcher must undertake to accomplish qualitative research, access to a safe, private space also lent itself to the collection of richer data. [55]

In the current study, the data obtained from my interviews with incarcerated men were generally deeply personal and detailed, which I believe further reflects the advantages of having a private, distraction-free interview setting (ECKER, 2017). Although I designed the interview questions similarly for each field site, the more private setting of the prison permitted participants (and myself) to reflect on questions and answers, resulting in thick data that were rich with personal narratives. [56]

In contrast, the data from my interviews with men living on the street varied a great deal in quality, which was unsurprising given the unpredictability of conducting research on busy streets. The same interview style and structure thus yielded less consistently rich data. As a result, my data collection efforts in the field became more ethnographic in nature, relying on observations and unstructured conversations and less on an interview guide. Researchers speaking to the problematic nature of gatekeeping should thus be aware that the absence of gatekeepers in public, open spaces may facilitate easier access to hard-to-reach populations, but developing the type of conversation that yields the rich data that qualitative researchers seek can be much more difficult in these settings. Despite the long-winded negotiations that were necessary before I could begin collecting data at the Eagle Foundation, the narratives from these men proved immensely useful because—much like in the prison setting—I was able to obtain rich, first-hand accounts of their lives in a private, quiet interview location. [57]

Tweaking data collection methods is an inherent part of the grounded theory approach to qualitative research that I used in this study, and I was thus prepared for the change in my methods. However, qualitative scholars who are for whatever reason committed to a semi-structured or structured interview schedule should be aware that, despite the absence of gatekeepers, more rigid data collection methods may prove difficult on the streets. Overcoming the arduous process of accessing prisoners, on the other hand, can result in the reward of highly rich, deep data, despite the difficulties of conducting field research in restrictive prison settings. [58]

Finally, the quality of the data is also affected by the researcher's ability to validate what participants say. With the Eagle Foundation's denial of my request to access resident records, and with the complete absence of such records in the case of the men living on the street, I had little choice but to accept the truth of my participants' narratives in these settings. In contrast to my inability to validate any part of the narratives of the men that I interviewed on the street and in the Eagle Foundation, I was able to validate some of the data obtained from the prison interviews using the detailed records provided to me by the Department of Corrections. Although states in the U.S. vary in their willingness to approve prison research, scholars should bear in mind that—if granted approval—the formalized process of gaining and maintaining access in the prison system can result in more rigorous methods of data validation. It may thus be worth researchers' time and effort to surmount the admittedly arduous task of gaining initial access to prisons, especially because doing so could reinvigorate the important work of "getting inside and around" (WACQUANT, 2002, p.371) prisons. [59]

In this article, I have argued that researchers should acknowledge both the challenges and the benefits of gatekeeping when studying vulnerable populations. In addition to the empirical benefits of comparing populations that differ in their vulnerabilities, researchers can draw on comparative qualitative research to develop a more critical and informed understanding of how gatekeeping affects the research process. It is important to reiterate that the purpose of this article is not to negate the very real emotional (LIEBLING, 1999) and practical (BOSWORTH et al., 2005; SCHLOSSER, 2008; UMAMAHESWAR, 2014) difficulties of conducting prison research; rather, my argument is that scholars must not, in their discussions of these difficulties, overlook some of the benefits offered by prisons as field sites. [60]

In this study, I quickly found that, despite the ease of accessing persons living on the street (ANDERSON & CALHOUN, 1992), studying homeless persons in this way was very complicated. Although I was able to approach the men without negotiating first with multiple gatekeepers (as I had to do in both the prison and the Eagle Foundation), approaching potential participants in this way raised a host of ethical and methodological questions. The spontaneous way in which my interaction with the participants on the street began made it apparent very quickly that written informed consent was difficult to administer when interviewing on the street. Additionally, researchers studying homeless persons on the street should expect to confront difficult decisions about how (if at all) participants should be compensated for their participation. Finally, they should be prepared to develop a principled way of deciding whether (and how) to include participants under the influence of drugs and/or alcohol. [61]

Perhaps most significantly, relying solely on the data obtained from interviews with men on the street would have been very problematic. First, homeless men living on the street represent a unique subset of the homeless population. More importantly, as mentioned, the data obtained in this setting varied greatly in depth and richness because interviewing in public, open spaces was very distracting and (in some cases) dangerous. For this reason, the project benefited greatly from the data collected at the Eagle Foundation. Although it took a long period of negotiation, once I gained permission to conduct research at the facility, I was able to interview residents in a secure setting for an extended period. Compromising with gatekeepers at the facility thus resulted in a safe, private research location that proved very helpful in obtaining rich data that is harder to obtain in the street. [62]

Conducting prison research is no doubt challenging in ways that prior research (SCHLOSSER, 2008; UMAMAHESWAR, 2014; WACQUANT, 2002) has documented. In this study, negotiating access to the prison took months on end. Even after gaining access, the prison regimen often interfered with how easily I could conduct the interviews. For instance, the interviews had to be scheduled at particular times so as to minimize disruptions to the prison routine, and there were frequently long waits before I was able to meet with the participants. I have argued, however, that researchers can and should acknowledge and exploit the (perhaps unintended) benefits of the rigorous gatekeeping that characterizes prison research. The informed consent process was much more predictable and consistent in the prison setting and the prison staff typically arranged a quiet, private area in which interviews could be conducted—a particularly helpful feature of prison research for reasons related both to participant confidentiality and data quality. Finally, although conducting interviews with incarcerated persons can be mentally and emotionally taxing, the very rules that can frustrate researchers seeking a completely independent research process can assist them in responding to participants' requests for assistance outside the scope of the research project. [63]

The difficulties that researchers face in different research settings are no doubt contingent on researchers' own personal and demographic variables. I believe, however, that my experiences with conducting research in settings characterized by varying levels of gatekeeping are valuable for researchers across the demographic spectrum. Considering the emphasis qualitative researchers have placed on the power of gatekeepers to inhibit prison research, it is simultaneously important to keep in mind the complexities involved in conducting research in the absence of gatekeepers. [64]

1) Readers interested in issues related to informed consent in the prison setting may review SCHLOSSER (2008) and UMAMAHESWAR (2014). <back>

Anderson, Leon & Calhoun, Thomas C. (1992). Facilitative aspects of field research with deviant street populations. Sociological Inquiry, 62(4), 490-498.

Blumstein, Alfred & Beck, Allen J. (1999). Population growth in U.S. prisons, 1980-1996. Crime and Justice, 26, 17-61.

Bosworth, Mary; Campbell, Debi; Demby, Bonita; Ferranti, Seth M. & Santos, Michael (2005). Doing prison research: Views from the inside. Qualitative Inquiry, 11(2), 249-265.

Charmaz, Kathy (2003). Qualitative interviewing and grounded theory analysis. In James A. Holstein & Jaber F. Gubrium (Eds.), Inside interviewing: New lenses, new concerns (pp.311-330). Thousand Oaks, CA: Sage.

Duke, Karen (2002). Getting behind the "official" line: Reflections on dilemmas of access, knowledge and power in researching policy networks. Journal of Social Policy, 31(1), 39-59.

Ecker, John (2017). A reflexive inquiry on the effect of place on research interviews conducted with homeless and vulnerably housed individuals. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 18(1), Art. 5, http://dx.doi.org/10.17169/fqs-18.1.2706 [Accessed: June 21, 2018].

Giallombardo, Rose (1966). Society of women: A study of a women's prison. New York, NY: John Wiley & Sons.

Goffman, Erving (1961). Asylums: Essays on the condition of the social situation of mental patients and other inmates. Garden City, NY: Anchor Books.

Gowan, Teresa (2002). The nexus: Homelessness and incarceration in two American cities. Ethnography, 3(4), 500-534.

Hagan, John & Dinovitzer, Ronit (1999). Collateral consequences of imprisonment for children, communities, and prisoners. Crime and Justice, 26, 121-162.

Heffernan, Esther (1972). Making it in prison: The square, the cool, and the life. New York, NY: Wiley-Interscience.

Kiddey, Rachael & Schofield, John (2011). Embrace the margins: Adventures in archeology and homelessness. Public Archeology, 10(1), 4-22.

Liebling, Alison (1999). Doing research in prison: Breaking the silence?. Theoretical Criminology, 3(2), 147-173.

Liu, William Ming; Stinson, Ren; Hernandez, Jovan; Shepard, Samuel & Haag, Sarah (2009). A qualitative examination of masculinity, homelessness, and social class among men in a transitional shelter. Psychology of Men & Masculinity, 10(2), 131-148, http://citeseerx.ist.psu.edu/viewdoc/download?doi=10.1.1.536.2519&rep=rep1&type=pdf [Accessed August 22, 2018].

Metraux, Stephen; Roman, Caterina G. & Cho, Richard S. (2008). Incarceration and homelessness. Paper for the 2007 National Symposium on Homelessness Research, Washington, D.C., March 1-2, 2007, https://www.huduser.gov/portal//publications/pdf/p9.pdf [Accessed: June 21, 2018].

Milne, Catherine (2005). Overseeing research: Ethics and the Institutional Review Board. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 6(1), Art. 41, http://dx.doi.org/10.17169/fqs-6.1.531 [Accessed: August 15, 2018].

Newburn, Tim & Stanko, Elizabeth A. (1994). Just boys being boys?. Men, masculinities, and crime. Abingdon, Oxon: Routledge.

Passaro, Joanne (1996). The unequal homeless: Men on the streets, women in their place. New York, NY: Routledge.

Pawelz, Janina (2018). Researching gangs: How to reach hard-to-reach populations and negotiate tricky issues in the field. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 19(1), Art. 2, http://dx.doi.org/10.17169/fqs-19.1.2878 [Accessed: June 21, 2018].

Phelps, Michelle S. & Pager, Devah (2016). Inequality and punishment: A turning point for mass incarceration? The ANNALS of the American Academy of Political and Social Science, 663(1), 185-203.

Rabionet, Silvia E. (2011). How I learned to design and conduct semi-structured interviews: An ongoing and continuous journey. The Qualitative Report, 16(2), 563-566, http://www.nova.edu/ssss/QR/QR16-2/rabionet.pdf [Accessed: August 15, 2018].

Reeves, Carla L. (2010). A difficult negotiation: Fieldwork relations with gatekeepers. Qualitative Research, 10(3), 315-331.

Sabo, Don; Kupers, Terry A. & London, Willie (2001). Prison masculinities. Philadelphia, PA: Temple University Press.

Simon, Jonathan (2000). The "society of captives" in the era of hyperincarceration. Theoretical Criminology, 4(3), 285-308.

Snow, David A. & Anderson, Leon (1993). Down on their luck: A study of homeless street people. Berkeley, CA: University of California Press.

Stein, Judith A. & Gelberg, Lillian (1995). Homeless men and women: Differential associations among substance abuse, psychosocial factors, and severity of homelessness. Experimental and Clinical Psychopharmacology, 3(1), 75-86.

Taylor, Cheryl (1993). On the conduct of homelessness research: Lessons from a qualitative study of women diagnosed with chronic mental illness. Issues in Mental Health Nursing, 14(4), 425-432.

Toro, Paul A. (2006). Trials, tribulations, and occasional jubilations while conducting research with homeless children, youth, and families. Merrill-Palmer Quarterly, 52(2), 343-364.

Umamaheswar, Janani (2014). Gate-keeping and the politics of access to prisons: Implications for qualitative prison research. Journal of Qualitative Criminal Justice and Criminology, 3(1), 238-267, http://www.jqcjc.org/documents/v3i2.pdf [Accessed: June 21, 2018].

Van Geest, Jonathan B. & Johnson, Timothy P. (2002). Substance abuse and homelessness: Direct or indirect effects?. Annals of Epidemiology, 12(7), 455-461.

Ward, David A. & Kassebaum, Gene G. (1965). Women's prison—Sex and social structure. Chicago, IL: Aldine.

Western, Bruce & Wildeman, Christopher (2009). The black family and mass incarceration. The ANNALS of the American Academy of Political and Social Science, 621(1), 221-242.

Janani UMAMAHESWAR is an assistant professor in the Department of Sociology at Southern Connecticut State University. Her research interests are in the areas of gender, incarceration and punishment, the life course, and qualitative research methods.

Department of Sociology
Southern Connecticut State University
501 Crescent Street
New Haven, CT 06515, USA

Umamaheswar, Janani (2018). Studying Homeless and Incarcerated Persons: A Comparative Account of Doing Field Research With Hard-to-Reach Populations [64 paragraphs]. Forum Qualitative Sozialforschung / Forum: Qualitative Social Research, 19(3), Art. 24, http://dx.doi.org/10.17169/fqs-19.3.3053.

Forum Qualitative Sozialforschung / Forum: Qualitative Social Research (FQS)

ISSN 1438-5627

Creative Commons Attribution 4.0 International License