Ethical and Epistemological Issues in Intimate Insider Ethnographic Research on Crime and Deviance

Abstract

There is considerable literature on intimacy, and friendships developed in the field of social research. However, there is a lack of research exploring the ethical issues that arise when researching pre-established friendships. Subsequently, there is a lack of research exploring the ethical issues that arise through the intersection of ethnographic, intimate insider, and crime and deviance research. Intimate insiders are researchers who study in a community where they have pre-existing relationships. In this article, the author uses their study of the illicit streaming device market to illustrate the ethical challenges they encountered while conducting ethnographic research with pre-existing friends who were involved in crime and deviance. The article argues that this intersection raises unique ethical challenges around confidentiality, power imbalances, role conflict, the risk of harm, and deception, and describes how these issues were overcome. It concludes that although such research raises ethical issues, it also offers benefits.

Introduction

When ethnographers study pre-established friends who engage in crime and deviance, they engage in three research types: ethnographic, intimate insider, and crime and deviance. Ethnography is a methodological approach that involves participant observation as a core method, focusing on the study of people in their natural settings (Treadwell 2020). Intimate insider research occurs when the researcher has pre-existing relationships with participants prior to the commencement of the study (Taylor 2011). Crime and deviance research refers to studies in which participants engage in illegal or socially disapproved acts (Lawson and Heaton 2010). When these three types of research are combined, it creates unique methodological and ethical challenges that the researcher must address. This includes balancing the researcher's obligations to their research project with their friendship obligations to their pre-existing friends. In this article, I illustrate these challenges by reflecting on my experiences using ethnographic methods to research individuals with whom I had pre-existing friendships before starting the research, as well as their associates who were new contacts, who used or traded illicit streaming devices (ISDs) in towns in Northeast England. The lessons learned from this paper are applicable to any study that involves transforming a membership role (i.e., where the researcher is a member of their study community) into a research role in any context where the research involves active criminals.

I begin by reviewing the insider research, intimacy, friendship, and crime and deviance literature. I argue that there is a lack of research exploring the ethical issues that arise through the intersection of ethnographic, intimate insider, and crime and deviance research. Then, I describe the ethical issues I encountered while researching this intersection and explain how the existing literature was useful in mitigating some of these issues, as well as its limitations. I conclude by arguing that ethnographers encounter unique ethical issues when researching pre-existing friends who commit acts of crime and deviance. Such issues can be reduced if the researcher engages in reflexivity.

Literature Review: Researching Friends in Social Research

Traditionally, social researchers were discouraged from having or forming close relationships with participants due to the belief that they needed to distance themselves from those being studied to ensure objectivity. Since the reflexive turn in qualitative sociology, there has been a shift toward embracing intimate relationships (e.g., friendships) as long as researchers engage in continuous reflexivity because reflexivity helps researchers to produce ethically sound qualitative research (Berger 2015; Maier and Monahan 2009; Rahman and Deuchar 2024; Taylor 2011). Olmos-Vega et al. (2022, 242) define reflexivity as “a set of continuous, collaborative, and multifaceted practices through which researchers self-consciously critique, appraise, and evaluate how their subjectivity and context influence the research processes.” This ideological shift has led to qualitative researchers forming intimate relationships with their participants, and the debate has shifted to determining how close researchers can get to their participants before they become too close (see Maier and Monahan 2009 for a review of this debate in the context of crime and deviance).

There is a growing number of studies on crime and deviance in which ethnographers have embraced this position. To overcome the challenges in gaining access to criminal and deviant groups, qualitative researchers often study communities where they are insiders and develop friendships with participants while in the field (Goffman 2015; Pearson 2012) or they research people with whom they have pre-existing friendships (Adler 1993; Contreras 2012; Cristiano 2023; Taylor 2011). Field-based friendships and insider research are different from intimate insider research. Field-based friendships are when the researcher develops friendships with their participants while in the field. An insider is a researcher who belongs to a group due to a shared characteristic (e.g., culture, race, or profession) (Adler and Adler 1987). Intimate insiders are researchers who study people with whom they were friends before they began their research (Taylor 2011, 8). When researchers develop friendships in the field or have pre-existing friendships with their participants, they must reflect on how this relationship impacts their research.

A comprehensive body of literature exists that describes the advantages and disadvantages of insider social research and friendships developed in the field. As insiders are close and familiar with their community of study, they can easily gain access to hidden populations and encourage them to open up, thus enabling greater depth to the data gathered because they have already gained their trust and established rapport (Brewis 2014; Chavez 2008; Dwyer and Buckle 2009; Ergun and Erdemir 2010). Insider researchers also possess pre-existing knowledge of the community, which enhances communication between researchers and participants (Chavez 2008).

Hidden populations such as active criminals are often difficult to access in their natural settings; consequently, ethnographers often utilize their social networks to gain access to this study population, resulting in a snowball referral effect (e.g., Adler 1993; Bourgois 2003; Hall and Antonopoulos 2016). Without intimate insider access, such studies would often not be possible.

Insider research also has some drawbacks, one of the main ones includes insider bias; researchers may be too close to their participants and, subsequently, engage in selective reporting, or their analysis may misrepresent them to portray them in a positive light (Chavez 2008). This closeness also risks “going native,” which involves ethnographers becoming too involved in the community under study, adopting their participants’ worldviews, and thus losing objectivity and distance (O’Reilly 2009). When researchers have intimate relationships with their community, they may feel pressured to engage in community activities, which can include lawbreaking, raising legal issues (Adler 1993; Bucerius 2013; Cristiano 2023; Goffman 2015; Pearson 2012). This is related to the social contract to show solidarity with friends. Goffman (2015) is a pivotal ethnographic case demonstrating this issue. She took an ethnographic approach to policing in a poor urban neighborhood in West Philadelphia, befriended her participants, and went “native.” She was accused of participating in a serious crime (i.e., conspiracy to commit murder) but was not prosecuted (Lubet 2015). She was defended (Volokh 2015) and criticized (Lubet 2015) by the academic community due to her questionable ethics. Venkatesh (2008) is another ethnographic example of when a researcher engaged in crime with their participants by becoming a gang leader for a day. They were studying drug dealers, and they befriended their key informant, who was a gang leader. This pressure to engage in law-breaking increases if the researcher adopts a covert approach, as they attempt to prove they fit in with and become a member of the community under study (Adler 1993; Calvey 2017). Some studies would not have been possible if the researcher had not participated in community activities (e.g., lawbreaking with their participants) (Adler 1993; Moretti 2019; Pearson 2012).

As insider researchers are familiar with the community of study, they may overlook important information that an outside researcher might not, and participants may assume the researcher is an expert on the study topic and subsequently not explain information in detail (Bucerius 2013; Dwyer and Buckle 2009). Another main challenge when engaging in qualitative research as an insider, is experiencing a role conflict between the researcher role and friendship role. This includes the risk of reporting information beyond that shared during the research period, and/or beyond what is directly related to the report (Abeyasekera 2017; Brannick and Coghlan 2007; Brewis 2014; Cristiano 2023; Taylor 2011). Role conflict is explored further later.

As identified by Taylor (2011), there is extensive literature on insider research, intimacy, and friendship in social research. It considers the methodological or ethical challenges such research poses (Brewis 2014; Chavez 2008; Dwyer and Buckle 2009; Ergun and Erdemir 2010). However, it primarily focuses on friendships that develop during fieldwork, when informants become friends, there is a lack of studies describing the ethical challenges ethnographers encounter when researching pre-existing friends (Abeyasekera 2017; Adler and Adler 1987; Contreras 2012; Cristiano 2023; Irwin 2006; Taylor 2011). Such research is important because when researchers have pre-established relationships with participants, this creates additional challenges compared to friendships with participants that develop in the field (Brewis 2014; Irwin 2006; Taylor 2011). Challenges such as how researchers should behave when in the field and how to write about the field, these issues arise due to the power imbalance in the field, which is in favor of the researcher (Irwin 2006, 159). These issues are explored later in this paper using my doctoral study as an illustration.

There is an extensive body of literature that provides guidance or empirical accounts of the ethical challenges that arise when using qualitative methods to research crime and deviance (for a review of this literature, see British Society of Criminology 2015; Calvey 2017; Cowburn et al. 2016; Treadwell 2020; Yates 2004). However, the methodological and ethical issues that arise through the intersection of ethnographic, intimate insider, and crime and deviance research have been under-examined. Subsequently, there is a lack of guidance for researchers engaging in such research. Some studies have explored the ethical issues that arise when engaging in ethnographic research with pre-existing friends who engage in illicit drug markets (Contreras 2012; Cristiano 2023; Taylor 2011). Both Contreras (2012) and Taylor (2011) identified power relationships and role conflict as the two main ethical issues of such research. Similarly, Cristiano (2023) also identified these issues alongside additional ones, including confidentiality, informed consent, and legal harm. Cristiano (2023) provided the most recent and comprehensive account of the ethical issues arising due to the intersection of intimate insider, ethnographic, and illicit drug research. Some of the lessons learned from those studies are specific to researching pre-existing friends involved in illicit drug markets rather than being applicable to any study that involves researching pre-existing friends in any context where the research involves active criminals.

In this article, I present a new reflexive account of intimate insider, ethnographic research, with active criminals, offering new insights to the literature by discussing how this type of research creates issues around confidentiality, power imbalances, role conflict, the risk of harm, and deception. I contribute to the field of methodological research on crime and deviance by providing an account of the implications of power imbalances and how I navigated these. I also contribute by providing an account of the ethical challenges that arise in crime and deviance studies when working with both pre-existing and new relationships and juggling overt and covert roles with different people in the same community of study. I discuss how I navigate these issues. I draw on my study of the ISD market to illustrate the ethical challenges researchers may encounter when conducting ethnographic research with pre-existing friends and their social networks involved in crime and deviance.

Methodology

This is a reflexive paper that I wrote by drawing on field notes from my doctoral study, which included observations and conversations with research participants as well as emerging methodological and ethical issues. This section provides an overview of the research on which this reflexive paper is based. This article originates from ethnographic fieldwork with two separate communities of ISD buyers and sellers across two towns, supplemented by interviews with participants in another two towns. My sample also included two types of end-users: about twenty patrons watching illegally streamed Premier League sports in public houses and about sixty household users watching illegally streamed content (including movies, television shows, and Premier League sports) in private homes. It also included interviews with two pub landlords who illegally streamed sports in their pubs. Some users were aware they were illegally watching streamed content, while others were not. It is not possible to give precise numbers of participants because, in addition to formal interviews with some participants, I also collected data through informal conversations, participant observation, and second-hand knowledge from pre-existing friends. Sometimes, whether participants were discussing the same person or a new one was unclear. I also had knowledge of twenty-one ISD traders. These populations overlapped because they were often the same people.

My participants’ ages ranged between twenty-two and seventy. Most participants were men of a similar age or older than me, and the research settings were male-dominated. As a young woman researcher, I encountered fieldwork challenges related to my gender, which are explored in Potter (2024). My participants and I identified as being “White British” and coming from a “working-class background.” Participants described themselves as having a secondary school level of education.

Within the ISD retail market that I studied, there are traders who sell directly to customers who purchase for household use, buyers who purchase ISDs for home use, and household users who watch ISDs in their private homes. The buyer and end-user groups overlap. All buyers I spoke to were users, but not all users were buyers. There were always more end-users than there were buyers - watching sports, films, or TV shows is often a communal experience. See Potter (2024) for an introduction to the ISD retail market and the actors involved.

There is a global market for ISDs. An ISD is an internet-enabled device, such as an Android box or stick, that has been illegally modified to provide access to subscription-based content, such as Sky Sports channels, for free without paying subscription fees. Streams are illegally accessed via pirated applications. The device is inserted directly into television sets. The device itself is legal, but it becomes illegal when pirated applications are installed that provide unauthorized access to streaming services. The trade of ISDs and accessing illegal streams using pirated applications on streaming devices constitutes copyright infringement in the United Kingdom, Europe, the United States of America, and Canada (European Union Intellectual Property Office 2019). The victims of illegal streaming include the Premier League. Commercial premises, such as public houses, in the UK illegally broadcast Premier League matches if they do not have a commercial subscription from Sky Business or BT Business (Intellectual Property Office 2023). Even though the trade and use of ISDs is illegal, this behavior was normal within the communities of study. Illegal behavior, such as the trade of illegal goods and services, is usually hidden within society (Beckert and Dewey 2017).

As ISDs are often used in the living room, this has enabled digital piracy to become a crime engaged in with family and friends, which “normalises” it subsequently, some users consider using ISDs to be a socially disapproving act rather than a criminal act (The Industry Trust 2016, 4). Digital piracy is regarded as unacceptable at a societal level. However, trades involving a legal-illegal crossover, such as the trade of ISDs, elicit greater public tolerance and social acceptance than trades without a parallel legal market (Von Lampe 2016). The ISD trade is a form of crime and deviance.

The trade of illicit products, such as ISDs, is a hidden activity; therefore, ISD buyers, sellers, and users constitute hidden populations. One of the main challenges when researching hidden populations is identifying and gaining access to willing participants (Adler 1993). This issue was overcome because I had pre-existing social relationships with some participants. As part of everyday life, I frequented two public houses (“Stan’s Place and ‘Joe’s Tavern”) in Northeast England on weekends and home team football match days. I found that ISDs were being sold within these pubs, which provided an opportunity to research these markets, activities, and trading practices. Formal interviews and observations began after ethical approval was granted by my home institution (code FL18186).

As part of my doctoral study, I hung out in two different public houses observing the sale of ISDs by covertly observing how this transaction was carried out and how traders and their customers interacted during this transaction. I did not observe public house staff or patrons when ISDs were sold unless they were involved in this transaction. Observing the sale of ISDs offered first-hand and previously unavailable evidence of how traders advertised and sold their products to customers. It was not possible to inform those involved in the sale of ISDs that they were being observed as I did not know when the sale would occur. Blending in is essential when conducting covert research (Bulmer 1980; Calvey 2017), therefore, I sat with people from my social group in each pub, which allowed me to covertly observe pub traders without arousing suspicion.

To expand my sampling pool, I attended these pubs on special occasions and went to house parties with some participants. I identified and gained access to respondents via my social networks, which led to a snowball referral effect. Participants were willing to speak to me because I was already part of the research setting and had already gained many potential participants’ trust.

Ethnographic studies require researchers to choose between taking a covert or overt approach, each posing its own practical and ethical challenges, particularly when studying active criminals (Adler 1993; Calvey 2017). As a result, I used a combination of these approaches to balance these challenges (this is explored later). The study population can be divided into two groups. Those I had pre-existing relationships with, which included friends and acquaintances (people I knew but were not friends with), and those who were new contacts, some of whom became new friends, while others remained strangers. I took an overt approach with friends and a covert approach with acquaintances and new contacts unless it was safe to take an overt approach with them. With those whom I took an overt approach with, I provided them with a participant information sheet and used the practice of verbal consent (explored further later).

I did not record my relationships with participants in my fieldnotes. I relied on memory; therefore, I cannot provide precise numbers of how many participants I had pre-existing relationships with and how many were new contacts. This was a strategy I implemented to protect participants' external confidentiality in case law enforcement confiscated my fieldnotes.

Participants were interviewed in public houses or cafes, and interviews were guided by a topic guide. I asked traders about the mechanics and structures of the markets in which they operated. I asked end-users about their experiences of using ISDs. I asked both groups about their perceptions of risk, and the risk management strategies (if any) employed.

There was no alternative method that could have been used to research these topics other than trading myself. Trading ISDs is illegal and poses the risk of prosecution, therefore, this method was avoided. All participants that I formally interviewed were asked whether I could audio record the interview; if not, I took paper-based notes of key points. I wrote up detailed fieldnotes of interviews and observations, summarizing events and my initial reflections on them, as best I could remember, in an anonymized format. Pseudonyms were used for names and places to help preserve confidentiality.

I was an intimate insider among my community of study. Adler and Adler (1987) identified three membership roles of ethnographers: (a) peripheral member researchers, who refrain from engaging in the core activities of group members; (b) active member researchers, who participate in the core activities of the group without fully committing themselves to the members’ values and goals; and (c) complete member researchers, who are already members of the group or who become members during the research and engage in their core activities. Based on their typology, I was a peripheral member. I was part of the social group as I regularly attended events where people watched sports being illegally streamed, but I did not participate in the group’s core activities (e.g., illegal market activity). Similarly, Adler (1993) was also a peripheral member. She was part of the social group but did not participate in their core activities (e.g., drug smuggling).

Intimate Insider Research on Crime

In the following sections, I describe the main ethical issues I encountered due to engaging in ethnographic research with pre-existing friends, and their associates who were engaged in crime and deviance. I also describe how I minimized these issues. I organize these issues into five main categories: confidentiality, power imbalances, role conflict, the risk of harm, and deception.

Confidentiality

Confidentiality protects the privacy of research participants. It is important to most social research, but particularly when researching people engaged in crime and deviance, such as ISD traders and users, as a breach of confidentiality may result in participants experiencing emotional distress and prosecution. Research presents two types of confidentiality that a researcher must adhere to: external and internal (Tolich 2004). External confidentiality is breached when participants are identifiable in research reports to those outside the study. Internal confidentiality is breached when participants can identify others in the study by reading the final report. Goffman (2015) is an example of a breach of external confidentiality. Arguably, she was not vigilant enough in her efforts to ensure external confidentiality as people (e.g., journalists) outside her study could identify her participants and places of study after reading her ethnography (Manning et al. 2016). This may have caused emotional distress to the participants, as Goffman (2015) made assurances of confidentiality to them. Ellis’s (1986) ethnographic study of an isolated fishing community is a heavily cited example of when internal confidentiality was breached. In the study, Ellis disclosed secrets about her participants without considering whether it would cause them emotional distress. When she revisited the community (Ellis 1995), her participants raised concern that as they could identify themselves in her book, even though pseudonyms were used, others in the community would also be able to identify them. She found that this had caused emotional distress to her study community and had put a strain on the researcher–participant relationship. External confidentiality is a fundamental aspect of all ethical guidelines, but internal confidentiality is sometimes overlooked by researchers (Cristiano 2023, 6).

Kaiser (2009) advised that in qualitative studies that use a snowball sampling approach, such as my doctoral study, participants would be able to identify themselves and those they knew in the written report unless measures were implemented to protect their internal confidentiality. When I began data collection, I was initially concerned about internal confidentiality; however, the ethics literature lacked guidance on this issue. Subsequently, I discussed this with my research supervisors. They advised that it was not an issue in my study unless it had the potential to cause harm to participants or others, which it did not. Unlike other studies of illegal behavior (Cristiano 2023), my participants did not hide their involvement in illegal activity from other community members (e.g., other participants) because they were part of communities where crime and deviance was normal. Participants explicitly informed me they were not concerned if other members of the community that I was speaking with found out what they had discussed with me during interviews, explaining that crime and deviance is normal within their community. This is exemplified by the following participant, and other participants used similar phrases. “I’m not bothered if anyone finds out what I’ve said to you, I use dodgy boxes, so who cares everyone does it around here” (Broady, ISD user). Some participants informed me that they had no intention of reading the final report, which further reduced my concern. Subsequently, minimal efforts (e.g., using pseudonyms) were made to protect their internal confidentiality.

Protecting my participants' external confidentiality was my priority. After reading the participant information sheet and completing the verbal consent process, none of my participants expressed concerns about participating in my study or about how their data would be used. However, some participants expressed concerns about protecting their identity from people outside their communities. Some participants expressed concern that the police or Trading Standards would discover they were involved in illegal activity due to their participation in my study, which could result in prosecution. These participants asked me to promise that their identity would never be revealed to anyone outside their community.

When researching crime and deviance, there are limits to confidentiality. The Economic Social Research Council (ESRC) (2015, 24) guidelines state that researchers have a legal obligation to break confidentiality and inform authorities about criminal behavior if there is a risk of harm to other people or interviewees. Subsequently, when researching illegal activities, the researcher cannot guarantee that the participant’s confidentiality and privacy can be maintained (British Society of Criminology 2015; Elliot and Fleetwood 2017; ESRC 2015). This was explained to my participants verbally and in their participant information sheet. There is a debate about whether researchers should proactively disclose knowledge of illegal activity to law enforcement to aid them in their policing efforts (Elliot and Fleetwood 2017). During my study, I did not inform law enforcement that I had knowledge of criminal behavior because this would have been unethical, a breach of confidentiality, and would have caused other problems (explored further in Potter and Potter 2021). I was also not legally obligated to do so, as my study posed no risk of harm to other people or the interviewees.

As I researched individuals engaged in illegal activity, there was a risk that I might be asked, or even forced, to hand over my data to law enforcement (in the UK) under the Investigatory Powers Act (2016), as one researcher has been (Garrett 2014). This point is explored further later. I explained this to my participants and assured them that it was unlikely to happen, as researchers are rarely compelled to hand over their data to law enforcement (Elliot and Fleetwood 2017), however, I could not guarantee it would not happen. Everyone I asked was willing to participate in my doctoral study, suggesting they were not concerned about these risks.

To protect participants' confidentiality, as soon as the fieldnotes were typed up and the interviews had been transcribed, all participants, as well as the people and places they mentioned, were assigned pseudonyms. Identifying features were not recorded in fieldnotes as far as possible and were removed from the data during transcription. I also implemented other strategies to protect my participants' identities from outsiders by protecting my information security. I learned these strategies from personal communications with criminological ethnographers; these strategies are legal. Kindynis and Fleetwood (2024) outline some general principles of information security and their potential applications to ethnographic research on crime and deviance.

Consent

To protect the identities of my participants, I employed a practice of verbal consent. When conducting research with people in the social sciences, a key ethical requirement is to obtain the participant’s consent (ESRC 2015). All participants in my doctoral study were provided with a participant information sheet that described the study aims and informed them of their right to withdraw from the study. Thankfully, no one did.

It was expected that, as other ethnographic studies of crime and deviance have found, some participants (traders or users) would not provide written consent (Coomber 2002; Ensign 2003; Hobbs 2000). The ESRC Framework for Research Ethics states that when studying criminal and deviant groups; “. . . In some circumstances (such as when users of illegal drugs and illegal groups are involved) written consent might create unnecessary risks for research participants” (ESRC 2015, 31).

The use of signed consent forms may comprise issues of confidentiality and anonymity which are important issues where participants need protection. Participants may fear that signed consent forms may make the information they provide traceable to them which may make them vulnerable to potential investigation and prosecution by the criminal justice system (in the case of illegal activities) (ESRC 2005, 17).

As such, verbal consent was gained from ISD traders, buyers, and users, and pub landlords for interviews and some observations. The practice of verbal consent is commonly used in ethnographic studies when researching illegal activities (Coomber 2002; Ensign 2003; Hobbs 2000). After going through the verbal consent process, I found that everyone I spoke with was willing to participate in my study. Some expressed reluctance to sign any forms in case law enforcement saw these forms, which I assured them was unnecessary.

Power Imbalances

All social researchers must consider how their authoritative position as researchers may cause power imbalances in the field. Additional issues may arise depending on the study population. In my doctoral study, there was the potential for a power imbalance between the participants and the researcher for two reasons. The first issue was that I was researching friends and marginalized (e.g., socio-economically disadvantaged) groups. I was in a privileged position as a university-educated researcher who was studying marginalized groups with a secondary school-level education. I had a higher educational level than the participants, which had the potential to cause a power imbalance. Marginalized groups are vulnerable to being influenced by people in power (Adley et al. 2024). Most participants did not fully comprehend what a PhD degree was; some assumed I was studying to become a medical doctor. Bucerius (2013), in her study of drug users, also encountered this issue among her non-university-educated participants. After I informed my overt participants that I was studying a criminology research degree, some thought I was studying to become a crime scene investigator (CSI). “Oh, so you wanna be just like them on the television show CSI” (Lee, ISD trader). They were receptive to participating in my study when I clarified and explained that I intended to write a book about their involvement in the “dodgy box and stick market” (e.g., the ISD market). As my participants were unfamiliar with the concept of a dissertation, I explained that I was writing a book on ISDs as part of the requirements of my university degree. I interpreted writing a doctoral thesis as being similar to a book, in terms of the best way to describe it to potential participants, as they have a similar word length.

It has been widely identified that when a researcher studies their friends, this can create a power imbalance between the researcher and participants (Berger 2015; Irwin 2006; Taylor 2011). When researching friends, it is difficult to identify whether they are participating in the study because they genuinely want to or feel obligated to help their researcher-friend (Abeyasekera 2017; McConnell-Henry et al. 2009). Some friend-participants felt socially obligated to participate in my research, as exemplified by Tom (ISD user): “We’re mates right, so I wanna help you get your degree and go on to do great things.” Other friend-participants did not feel socially obligated to participate in my research. They inquired whether they would be financially compensated for their time, and some male participants hinted at whether they would be compensated with sexual favors or a romantic date. Women are often asked this question when they research men (Gailey and Prohaska 2011; Harries 2016; Ramogwebo et al. 2024). I responded that they would not be compensated for their time and encouraged them to participate by describing how the research would benefit my career and provide them with an opportunity to discuss their experiences of trading or using ISDs. My doctoral study was self-funded; therefore, I did not have the money to compensate my participants financially. All participants whom I asked to participate in my study were willing to even though they were aware they would not be compensated for their time.

Due to these power imbalances (e.g., friendship relationships and authority positions), I worried that participants may have felt pressured to participate in my study. To minimize these issues, I reiterated throughout the fieldwork that participation was voluntary, and if they decided not to participate or withdraw from my study, this would not affect my friendship with them.

Insider research can cause another type of power imbalance in the field.

In insider research, the researcher is knowledgeable about the community and/or phenomenon being studied, and thus the participant is not by default the ‘expert.’ This type of research might thereby be a disempowering experience for participants (Cristiano 2023, 8).

I had not considered this ethical issue when I began fieldwork, but it emerged as an unexpected issue during the fieldwork. When I formally began researching the ISD market and asked participants to share their experiences of the market, some stated, “You will know more about it than me; you’re doing a study on it, [and] you’ve been chatting to everyone” (Elisha). Another participant stated, “You’ll already know this; there's no point in telling you” (Alice). On another occasion at Stan’s Place, one friend-participant was considering purchasing an ISD. As they knew I had been interviewing ISD users, they asked me about the benefits of using them rather than asking their friends, who were ISD users. In response, another participant said, “Why are you asking her? She hasn’t got a dodgy box, I have!” Their tone of voice and facial expressions clearly indicated they were offended. To make the participants feel empowered, I responded, “Yes, ask them; they’ll know more than me.” These scenarios demonstrate that my insider knowledge of the ISD market made some participants feel disempowered.

Researchers should strive to empower their participants by increasing their control and participation throughout the study, thereby protecting their participants' emotional well-being. When conducting insider research, Taylor (2011) advises that to empower participants, the researcher must act naïve and unassuming about their knowledge of the study topic. In this context, acting naïve refers to the researcher pretending to have limited knowledge on the topic of study to elicit further information. When researchers are outsiders to the study population, they can act naïve without question (Bucerius 2013; Moretti 2019), but insiders are not afforded the same luxury. It has been recognized that acting naïve to elicit further information when researching people, especially active criminals, with whom the researcher has pre-existing relationships, is challenging (Contreras 2012; Cristiano 2023). This is because participants assume the researcher has expertise in the study topic due to their pre-existing personal connections with their participants and the illegal scene.

I also encountered similar challenges in my doctoral study. There were mixed responses among my participants when I acted naïve, with men being more receptive than women. This may have been due to gender dynamics. I was a young women researching men of a similar age or older than me. Male participants may have misinterpreted me acting naïve as flirting. Some, mostly men, were happy to elaborate and fill my knowledge gaps without question. Others, mostly women or participants who were often abrupt, suspected I knew this information and was “just messing with them for a laugh” (Vicky, ISD user). Another participant said, “Drop the act, smarty pants. I know you know this” (Rhianna, ISD user). Other participants, who were pre-existing friends, were offended that I had a lack of knowledge about the ISD market, as they thought I had not been listening to them, as they had spoken to me about ISDs before I started my doctoral study. I had been visiting the two research sites in a social capacity before I started fieldwork. On one occasion, when I asked to know more about a topic, one participant responded, “Seriously? Have you not been paying attention?.” When I adamantly claimed I genuinely did not know the information I was asking about, they questioned how I did not know this information after “hanging around” with them in the pub for such a long time. Similar situations happened several times with the same or different participants.

In response to my sceptical participants, as recommended by Contreras (2012), I explained to them that I used phrases like “tell me more” because I wanted to hear how they understood the local ISD market rather than writing about how I understood the market. At other times, I explained to participants that although we may have discussed this topic on a previous occasion as friends, I was revisiting it as a researcher to ensure the accuracy of the information. In a friendly way, I explained to participants, “I am only human, like everyone, I forget things.” This often neutralized the situation, and the participants forgave me. Some of my friend-participants explained that they forgave me because they trusted and cared about me, or perceived deception as being a tactic that researchers use to extract information from their interviewees. The latter is demonstrated by Vicky (ISD user), “I’ll let you off, mate, I know you are only doing your job.”

At times, I felt that I was deceiving participants by pretending to have limited knowledge about the ISD market when I began formal data collection with them, which made me feel uncomfortable, but it was a necessary part of the research process. I had conducted extensive literature-based research on the ISD market since its emergence, but, of course, they were unaware of this. Participants often directed me to relevant press releases on actors involved in the ISD trade who had been prosecuted. I appreciated their level of involvement in my study, even though I had already seen most of these publications. I was reassured by the literature that acting naïve is not deception; it is a research strategy to encourage participants to elaborate on their responses (Contreras 2012; Taylor 2011).

Role Conflict

As my doctoral study involved researching friends, I occupied the dual roles of friend and researcher. This dual relationship with participants presented a role conflict, which led to one main ethical issue. When researching friends, there is a risk that private conversations with participants may be recorded as research data and included in the final report. This invasion of privacy may cause the participants emotional distress if they read the report. Role conflict is a common issue identified in social research that involves researching friends (Abeyasekera 2017; Brannick and Coghlan 2007; Brewis 2014; Cristiano 2023; Taylor 2011). This issue arises due to two reasons. First, due to having pre-existing friendships with participants, there is a risk that researchers may unintentionally use pre-existing knowledge of participants in their written report of the study (Abeyasekera 2017; Brewis 2014).

Second, as the researcher performs multiple roles, the participants may forget that their friend is a researcher, confide in them, and disclose information they do not want to be recorded as research data (Brewis 2014). This may result in researchers experiencing feelings of betrayal and disloyalty when they disclose what participants said in a private setting to a public one, as they disseminate their data (Brewis 2014; Fraser and Puwar 2008). Role conflict is not exclusive to intimate insider research; it is common in ethnographic studies for participants to forget they are being researched due to the researchers' prolonged immersion in the field (Pearson 2012). However, the issue can be exacerbated when researching pre-existing friends, as participants must adjust to their friend in their new role as a researcher (Cristiano 2023). I encountered these issues in my doctoral study, which is expected since I used ethnographic methods to research participants with whom I had been friends with for several years prior to my doctoral study.

To minimize these issues, Taylor (2011) advises that the researcher must use their judgement and intuition to decide when their participants are confiding in them in a friendship capacity and when they are revealing data to a researcher. This is an essential skill researchers must develop when researching friends. When deciding what conversations and observational data to include and exclude from their study, researchers must balance the participants' right to privacy against the risk of misrepresenting them by excluding data to portray them in a favorable light. I found this a challenging skill to develop, as it is subjective; some information I considered insider knowledge, another researcher may not have. There are other factors that researchers must consider when deciding what data to include in their written report. Subsequently, as Brewis (2014) reports, researchers' accounts of their participants' social worlds in qualitative studies, particularly when researching friends, are not an accurate reflection of their world because some data have been omitted to protect participants' privacy.

I also employed a strategy recommended by Pearson (2012) to ensure that participants did not forget that I was a researcher, thereby avoiding the feeling that I had betrayed their trust. He advised that, if possible (e.g., when taking an overt approach), ethnographers should remind their participants throughout the research that the researcher is participating in activities or conversations in a researcher capacity rather than a friendship capacity. Consent in research is not a one-off process; it requires an ongoing discussion and renegotiation of trust throughout the research (Cowburn et al 2016). I subtly reminded my overt participants of my researcher status in the field. For example, when appropriate, I stated, “That is good data for my study.”

I also employed another strategy to minimize the risk of causing participants emotional distress due to including data they did not want to be included in my doctoral study, and to ensure the accuracy of the information. Several times during the data collection stage, I asked some of my participants (e.g., friends) if they would like to read my fieldnotes to offer feedback and remove anything they were uncomfortable with being made public. Member checking is a strategy for exploring the credibility of results and is recommended to all qualitative researchers, but it is considered essential when researching friends (Birt et al. 2016; Brewis 2014; Taylor 2011; Tillmann-Healy 2003). It was not possible to make this request to all my participants with whom I had taken an overt approach. This was primarily due to practical reasons (e.g., ethnographic studies are already time-consuming, Treadwell 2020) and partially because I was aware that when participants are requested to read the researcher's field notes or publication drafts, there is often a low response rate (Brewis 2014; Neal and Gordon 2001). These issues overlapped, as deadlines needed to be made, I could not wait for all participants to respond to my requests.

Risk of Legal Harm

In my doctoral study, I had guilty knowledge as I knew my participants were trading, or using ISDs. Ethnographic research with lawbreakers raises legal issues as researchers are subject to “guilty knowledge,” which involves directly witnessing or being informed about past or future crimes (Adler 1993). In many countries, there is no legal obligation to report crimes you are aware of to law enforcement, except in certain cases, such as those involving terrorism (Elliot and Fleetwood 2017). However, unlike journalists or solicitors, there is no confidentiality between researchers and participants. Researchers may be compelled to hand over their data to law enforcement, which could lead to their participants facing legal harm (e.g., prosecution). Under the legal system in England and Wales, police have the power to compel researchers to hand over their data to law enforcement under certain circumstances (see Elliot and Fleetwood 2017, 7–8 for a summary of this legislation). As participants were anonymized, even if law enforcement read my doctoral study or fieldnotes, it is unlikely they could identify them; therefore, they could not prosecute them.

If I had been forced to hand over my data, or reveal participants” identities to law enforcement and refused to then I might have received a prison sentence, as Scarce (1994) did. I discussed these concerns with my supervisors before data collection began, and they assured me that these risks were low; nevertheless, I was concerned about them, as they might result in either me or my participants being prosecuted, which would violate the ethical principle of non-maleficence (Murphy and Dingwall 2001).

These risks would have concerned any researcher but were particularly concerning to me as I was researching my social network. If participants were prosecuted, it would have resulted in irreversible damage to my relationships with my social networks and the study communities, in both my researcher and friendship roles. It would have caused both parties to experience emotional distress. Additionally, I worried that if my participants were arrested during my research, it might have resulted in harm to me, the researcher, as participants might have believed I had informed law enforcement about their illegal activities. See Potter and Potter (2021) for a discussion of how lawbreakers may react if they think the researcher has informed on them (e.g., with violence or threats of violence).

Deception

In my doctoral study, I used a combination of overt and covert techniques; the approach I took mostly depended on my relationship with the participant. I took an overt approach with friends and a covert approach with acquaintances and new contacts unless it was safe to take an overt approach with them. Using a combination of overt and covert approaches is common in ethnographic studies on crime and deviance, but researchers rarely reflect on the ethical challenges this approach poses (for exceptions, see Adler 1993; Moretti 2019). In his study of ticket touts, Moretti (2019) became a ticket tout. He adopted an overt approach with one trader, whom he named “Duck,” and took a covert approach with “Duck’s” associates. “Duck” advised Moretti not to reveal his researcher identity to his associates due to fears they would suspect he was a police informant and retaliate with violence. Intermingling approaches became a problem when “Ducks” associates discovered Moretti was a researcher, and “Duck” began to threaten Moretti on behalf of his associates. It was unclear whether Duck revealed Moretti’s researcher identity to his associates. Similarly, Adler (1993), in her study of upper-level drug dealing, was also vulnerable to her researcher identity being discovered by her covert participants as she took an overt role with key informants and a covert role with new contacts. I encountered similar risks in my study due to combining overt and covert approaches.

When possible, I aimed to take an overt approach, as I wanted to avoid the ethical concerns that covert research raises regarding deception and consent (see Bulmer 1980; Calvey 2017, for a discussion of the ethical problems covert research raises). It was not possible to inform all participants that I was researching them, mainly because the research took place in a public setting (e.g., a pub). Another reason was that I was advised by some friend-participants, who were regulars at Joe’s Tavern, that the patrons who did not know me might assume that I was a police informant and retaliate with violence or threats of violence.

Whatever you do, don’t tell that person you are doing research, he wouldn’t take it well. He will answer your questions if he doesn’t know he is part of your study, because he wouldn’t understand. He will think you’re a cop (Peter, friend and ISD-user).

Therefore, covert research was justifiable to protect the researcher from harm. This was not a concern at the other public house of study because Stan’s Place was not frequented by violent patrons, as Stan, the owner, was known to be violent toward disruptive patrons.

As my doctoral study involved mixing overt and covert roles within a small sample and within the same research settings, the main ethical issue it posed was a risk of harm to the researcher. There was a risk that my researcher identity might be discovered through unexpected disclosure between sample members. But, as discussed, this risk was unavoidable as it was not possible to inform all sample members about my study. Participants with whom I took an overt approach assured me that they would not inform people about my researcher identity if they believed those people might react confrontationally because they did not want me to be harmed.

Several self-identified strategies were implemented to hide my researcher identity among my covert study population. On my social media accounts, I did not have a university affiliation, or state that I was a researcher. I intentionally did not have a university website profile during the fieldwork stage of my research. Therefore, if a member of the public viewed my social media account or searched my name online, they would not discover my researcher identity. At conferences, I requested that conference delegates refrain from posting photos of me online. When taking a covert approach, other researchers could apply this strategy of avoiding having an online academic presence to hide their researcher identity. However, as having an online presence is a key part of being an academic (e.g., for networking and disseminating research), this approach may hinder career progression.

Conclusion

There is a paucity of literature providing a reflexive account of the ethical issues arising through the intersection of ethnographic, intimate insider, and crime and deviance research, as well as guidance on how to manage them (Contreras 2012; Cristiano 2023; Taylor 2011). Without such subject-specific guidance, researchers are required to review three distinct bodies of literature: one on the challenges of engaging in ethnographic research, another on the challenges of intimate insider research, and another on the challenges of crime and deviance research. This is problematic because as this article has shown, unique challenges arise when researching this intersection. Subsequently, the guidance available in the literature was not always helpful in my doctoral study for identifying and managing the challenges that arose due to researching this intersection. If more guidance were available, this would encourage more researchers to engage in qualitative, intimate insider research, on crime and deviance and to provide reflexive accounts of the challenges they encountered.

I extend this literature by drawing on my study of the illicit streaming device market to illustrate and raise awareness of the ethical challenges researchers may encounter when conducting ethnographic research with pre-existing friends, and their social networks who are involved in crime and deviance. Raising awareness of such challenges is important because if researchers enter the field unprepared, they may make mistakes that could damage their credibility and professional standing (Iphofen 2015).

In this article, I present a new reflexive account of intimate insider ethnographic research with active criminals. I contribute new insights to the literature by discussing how this type of research creates issues around confidentiality, power imbalances, role conflict, the risk of harm, and deception. I also contribute to the field of methodological research on crime and deviance by providing an account of the implications of power imbalances and how I navigated these. Other contributions include my account of the ethical challenges that arise in crime and deviance studies when working with pre-existing and new relationships and juggling overt and covert roles with different people in the same community of study. I discuss how I navigate these issues.

I conclude that ethnographers encounter unique ethical issues when researching pre-existing friends who are involved in crime and deviance. Arguably, the ethical issues encountered by most ethnographers (e.g., confidentiality and power imbalance) and long-term ethnographers (e.g., role conflict) are amplified when researching pre-established friends, and they are further complicated when researching active criminals. Balancing the dual role of friend and researcher is at the heart of most long-term ethnographic fieldwork, where building intimate relationships with people is central to gathering substantial data and gaining an in-depth understanding of the scene. The risks of legal and physical harm I described apply to anyone researching active criminals. I suggest that the ethical issues posed by intimate insider research can be mitigated if the researcher engages in reflexivity and acknowledges how their dual roles, as both friends and researchers, impact the research process.

Although intimate insider research raises ethical concerns, it also offers numerous benefits. It is often the case that without intimate insider access to the study population, research on active criminals, such as my doctoral study on which this paper is based, would not be possible, as trust is essential to encourage their participation. The lessons learned from this paper are applicable to any study that involves transforming a membership role into a research role in any context where the research involves active criminals.

I encourage more researchers to provide a reflexive account of the challenges they encountered when engaging in intimate insider research, in order to build a deeper understanding of the issues that such research poses. Such accounts can serve as guidance for other researchers. It is important that researchers study pre-existing friends in qualitative social research and provide a reflexive account of their experiences, as the researcher's relationship with their participants can significantly impact their ability to gain access to the study population and the type of data that can be collected. Therefore, valuable data could be missed if qualitative researchers only study strangers.