Youth participation in qualitative research: Challenges and possibilities

Abstract

Research often excludes youth participants, omitting their social and psychological realities, undermining their rights to participate and benefit from research, and weakening the validity of research. Researchers may be discouraged from including youth due to logistical (e.g. gaining access) or ethical (e.g. coercion risks based on developmental level) concerns. Increased discussion is needed around appropriate methods to use with child and youth participants that manage challenges related to developmental capacities, legal status, power differentials, and unpredictable aspects of qualitative research. This paper pools experiences of six researchers, describing solutions we have developed in studies employing varied qualitative methodologies with varied vulnerable youth subpopulations. We detail successful approaches to access, compensation, consent, assent, and confidentiality. Social work researchers are wellsuited to navigate the challenges, and we share our examples with the aim of facilitating increased youth participation in research.

Keywords

Youth participation qualitative research youth ethics young people human rights

Children and youth are often subjects of social work research, yet their voices are rarely heard directly. Most researchers collect data from parents, teachers, caseworkers or other proxy sources. Although research using proxies is informative, it represents research on children rather than with them and omits their first-person perspectives (MacNaughton and Smith, 2005). Children and youth have been considered incompetent, passive, conforming, immature, incomplete, and highly vulnerable subjects whose participation would be unreliable and susceptible to adult suggestion (Farrell, 2005; Fisher et al., 2013; Laenen, 2009; Mason and Hood, 2011; Mishna et al., 2004). Increasingly, however, researchers value children’s perspectives, seeing children as active agents in constructing and communicating their own realities (Kirk, 2007). Qualitative research, which seeks to understand and represent lived experiences and perceptions, is the ideal approach for understanding child and youth experiences.

Research with youth presents complexities, including logistical and ethical challenges. Social work researchers are highly qualified to navigate these challenges and contribute to the development of best practices in research with youth. Yet lack of guidance in the literature may contribute to the hesitation to involve youth in research. Mishna et al. (2004) reviewed ethical codes from psychology and social work in Australia, Great Britain, Canada, and the US, and found no discussion of research with children, apart from guidelines for vulnerable populations in general.

In this article, six researchers with varied research foci and qualitative approaches describe successful strategies for managing complexities associated with youth participation in research, to gain the benefits of first-person perspectives directly from youth. Our research focuses on several different US subpopulations of adolescents within the ages of 12–18 years old. We refer to them as “youth” throughout this article and draw on studies we completed recently. Lisa focused on youth transitioning out of the child welfare system (Schelbe, 2013). Amy’s study was with 12–14 year olds residing in domestic violence shelters (Chanmugam, 2014). The youth in Tally’s study were diagnosed with mental illness (Moses, 2009, 2011). Susan focused on youth identifying as lesbian, gay, bisexual, transgender and queer (LGBTQ) (Davis et al., 2009; Saltzburg and Davis, 2010). Lela’s research was with Mexican American youth experiencing dating violence (Williams, 2014). Joan’s study explored girls’ perceptions of fighting between girls (Letendre and Smith, 2011).

We begin by providing background information on the evolution of research with both children (individuals under the age of 18) and youth. We include literature that addresses child inclusion more broadly, because the discussion has not been divided by ages or developmental stages. Our article focuses more narrowly on youth because our study participants have been above age 12. While our article may inform qualitative research with children of varied ages, there may be considerations with younger children that we do not address (see Flores, 2008 for discussions on young child participants in research). After providing background, we review how qualitative research with youth differs from research with adults, followed by discussion of our approaches in four key areas: gaining access; collaborating with gatekeepers and data collection sites; managing informed consent and assent procedures and confidentiality considerations; and compensating participants. We share decisions made in studies using varied methods (e.g. focus groups, individual interviews, participant observation) with several US adolescent subpopulations. Our purpose is to assist scholars interested in qualitative research with youth by describing how we managed challenges, since considerations regarding methods are rarely addressed in the literature (Fleming, 2010). We conclude by highlighting social work researchers’ assets that make them especially well-prepared to conduct research with children.

Background

Children’s inclusion in research has increased exponentially in recent decades in social science and health disciplines (David et al., 2005; Duncan et al., 2009; Kirk, 2007; Laenen, 2009; MacNaughton and Smith, 2005; Rich and Ginsburg, 1999), especially in Europe, Australia, and Canada (Farrell, 2005; Kirk, 2007; Laenen, 2009; Mason and Hood, 2011). Several factors influence this trend: application of a human rights framework and child rights to participation (Fleming, 2010; Head, 2011; Kirk, 2007), empowerment and emancipatory research approaches (Fleming, 2010), social constructionist perspectives recognizing children as social actors shaping their environments, and evolving views on children’s capabilities, expertise, and the nature of childhood itself (Kirk, 2007; Mason and Hood, 2011; Rich and Ginsburg, 1999). In the 1990s, the US Congress expressed concerns about the inadequate knowledge base on children’s health since existing research focused on adults. Consequently, it directed the National Institutes of Health to increase child participation in basic, behavioral, and clinical research (Fisher et al., 2013), thereby addressing children’s rights to benefit from health research. This is a reflection of the growing recognition of the necessity of youth involvement in health behavior research since they have expertise on the motivations, beliefs, and contextual influences essential to understanding youth decision-making and health-related behavior (Rich and Ginsburg, 1999).

Fleming (2010) notes that the most common rationale cited for increased inclusion of children in research is their right to participate derived from a human rights perspective. The United Nations Convention on the Rights of the Child asserted children’s rights in the areas of protection, provision, and participation (Checkoway, 2011; Hart, 2008; Head, 2011; Kirk, 2007; Laenen, 2009; Mason and Hood, 2011; Mishna et al., 2004; Zeldin et al., 2005, 2008). Children’s participation rights include the right to have a voice in matters that concern them, rather than having their perspectives represented by others (Staller, 2008). In the US, a human rights approach to child and youth participation is considered innovative whereas in other countries child and youth participation is increasingly employed to strengthen communities and organizations (Zeldin et al., 2008) and improve research (Kirk, 2007; Mason & Hood, 2011; Mishna et al., 2004).

At the same time, protection rights, those that mandate adults to safeguard children from harm (Staller, 2008), shape the conduct of research with children, and have influenced the feasibility of research involving child participants in potentially unintended ways. In the 1990s, in conjunction with the mandate to increase child research participation, federal guidelines developed more explicit ethical guidelines for research with children, some of which serve to hinder research with children (Fisher et al., 2013). The US Department of Health and Human Services is currently considering regulation revisions for the first time in decades, and a task force convened by the Society for Research in Child Development developed recommendations to facilitate research with children by balancing their protection and participation rights (Fisher et al., 2013). For instance, child development researchers are questioning automatically classifying children as a “vulnerable” class in minimal risk research (Fisher et al., 2013). The taskforce recommendations include improvement of informed consent processes, reconsideration of the definition of risk, and reconsideration of the definition of emancipated minor and circumstance when guardian permission could be waived.

Evolving views on children’s capabilities and expertise have also influenced the trend towards greater inclusion of child and youth voices in research (Kirk, 2007; Mason and Hood, 2011; Rich and Ginsburg, 1999). Scholars note that children’s competence as research participants has now been recognized, whereas they were previously viewed as incompetent, passive, conforming, immature, incomplete, and highly vulnerable participants whose participation would be unreliable, susceptible to adult suggestion, and ultimately provide less legitimate knowledge (Farrell, 2005; Laenen, 2009; Mason & Hood, 2011; Mishna et al., 2004). Proxy reporters (e.g. parents, teachers) were traditionally considered more reliable reporters of child and youth experiences. Social constructionist frameworks influencing research have contributed to the idea that children and youth are active social actors capable of shaping the world around them, rather than objects (Kirk, 2007; Mason & Hood, 2011). Furthermore, they have expertise that can guide research, for example on how youth make health-related decisions (Rich and Ginsburg, 1999). In fact, as one scholar notes, adults reporting on experiences of children may misrepresent their experiences:

We now understand that adult representations and interpretations of children’s lives might say more about the observer than the observed, and, to avoid this, it has come to be seen as essential to convey children’s own accounts at first hand – to include their voices (Mullender et al., 2000: 3).

How is research with children different from research with adults?

The notion of children’s vulnerability in research is tied to actual and perceived differences in research with children compared to adults. Before delving into these, however, we note that differences in research with children versus adults may be overemphasized. In this vein, Kirk (2007) cites overlooked similarities regarding the need to: consider the broader societal context, use appropriate and rigorous methods, protect participants from harm, respect individuality, maintain confidentiality, manage disclosures, establish rapport, use clear questions, and avoid imposing one’s own views. In any case, differences considered particularly germane to research procedures include legal status, developmental capacities, and power differentials (David et al., 2005; Duncan et al., 2009; Kirk, 2007; Laenen, 2009; Mishna et al., 2004; Rich and Ginsburg, 1999).

Legal status and protection

Children have a different legal status from adults, and adults serve in protective roles for them. This is most directly reflected in the requirements for parental or legal guardian consent in addition to child assent, and possible interaction with other gatekeepers who manage access to children. This introduces logistical concerns not present with most adult participants (Alderson, 2005; Duncan et al., 2009; Kirk, 2007; Laenen, 2009). Also, possibilities exist for adults to deny participation opportunities of interested children or, conversely, to coerce unwilling child participants (Kirk, 2007). In qualitative research, concerns may be particularly heightened that children might talk about topics families prefer to remain private which may impact researchers gaining access to children and youth participants.

Developmental capacities

The emotional, cognitive, and language abilities of children differ from those of adults and vary considerably across children; researchers cannot assume that same-age children function at the same level (Duncan et al., 2009; Kirk, 2007). Researchers encounter variation in children’s competence to comprehend their rights, assent, and consent information; confidence to assert their wishes to adults (including a ‘deliberate objection’ to participation or an explicit wish to withdraw); and ability to predict participation aspects that might make them uncomfortable (David et al., 2005; Duncan et al., 2009; Mishna et al., 2004,). Given their limited life experiences (David et al., 2005), the research context may be unlike any experiences children and youth have had. As such, researchers must anticipate and be prepared to accommodate a variety of skills and comfort levels at every stage of the research process and to meet developmental capacities of youth.

Assessing for children’s competency, referring to individuals’ understanding of the basic purpose of research and what they will be expected to do, as well as their capacity to express clear preference, is inextricably tied to consideration of developmental capacities. On the one hand, there is a general consensus that there is no clear link between children’s chronological ages and their competency, and evidence that child’s competency is related less to age and more to life experiences (Alderson, 2005; Coyne, 2010; Kanner et al., 2004). Competency is multifaceted and includes considering interactions among a child’s age, maturity, cognitive ability, moral development, psychological state, and the social environment (Kanner et al., 2004). On the other hand, there are no clear DHHS or local IRB guidelines for this assessment or its documentation (Code of Federal Regulations 45 CFR 46:401–409, March 8, 1983); researchers are expected to use all of the available information to make this determination informally. Kanner et al. (2004) assert that as these considerations exist upon a continuum, and researchers must assess children as individuals rather than adopting arbitrary guidelines.

The National Commission for the Protection of Human subjects positions that while adolescents technically are children, “there are occasions when the rules that protect children are not necessarily appropriate for adolescents and may actually hinder the research process.” (Hester, 2004: 28). Unless there are severe cognitive, medical or psychological conditions, there is greater confidence (inferred) that adolescents are capable of understanding “the symbolic meaning of a signature” (Ungar et al., 2006: S32), and other important aspects of the assent process. It is the responsibility of the researcher to understand the cognitive development and abilities of the phases of adolescents to ensure that the study is developmentally appropriate (See more details about assessing competency in children in Ungar et al. (2006) and Hester (2004) in adolescents.). The changing view of both younger children and adolescents as social actors possessing competencies requires more debate and exploration about competency and guidelines pertaining to research (Coyne, 2010).

Still, researchers emphasize that it is feasible to design developmentally appropriate procedures despite varied developmental capacities, with some asserting that even younger children are often underestimated (Mishna et al., 2004). According to Flores (2008), age may inform how someone participates, but age should not determine whether someone participates. Suggestions for accommodating varied developmental levels include tailoring the assent process, for example by giving specific concrete information, followed by an assessment of youth understanding of the study and their rights (Alderson, 2005; Kirk, 2007; Mishna et al., 2004). Children need specific information on how to refuse or withdraw freely from participation if they prefer to (Kirk, 2007), with periodic follow-up throughout their participation on their willingness to continue (Kirk, 2007; Mishna et al., 2004).

Power differentials

Although qualitative researchers attend to issues of unequal power regardless of participant ages, the imbalance between adult researchers and child participants is particularly acute (David et al., 2005; Duncan et al., 2009; Eder and Fingerson, 2002; Kirk, 2007; Laenen, 2009; Mishna et al., 2004). Children are so often evaluated by adults, which can contribute to a sense of powerlessness in the research context (Laenen, 2009). In Laenen’s study, for example, youth brought previous negative experiences with adults in authority into the research situation, voicing suspicion of procedures like note-taking and audio-recording. If a child closely associates the researcher with an institution supporting the study, her or his perception of the researcher’s power (and by extension, their own lack of choice and agency) could be magnified further (Laenen, 2009; Mishna et al., 2004). Power differentials can influence every aspect of participation, including the potential for harm, consent, real or perceived coercion, and participants’ confidence to seek clarification, skip a question, or withdraw; they can also influence the child’s responses and trustworthiness of data (David et al., 2005; Duncan et al., 2009; Eder and Fingerson, 2002; Kirk, 2007; Laenen, 2009, Mishna et al., 2004).

Qualitative approaches provide advantages in negotiating power issues in research with children. These include flexibility in responding to varied literacy and assertiveness levels, methods inviting participant autonomy and voice, and the opportunity to understand how children think about and experience the world (Rich and Ginsburg, 1999). Other strategies for mitigating the power imbalances that are easily folded into qualitative research include encouraging collaborative, reciprocal interactions; for example, reciprocity in asking and answering questions reduces imbalances and conveys that the researcher is not looking for one “right answer” as teachers do in schools (David et al., 2005; Eder and Fingerson, 2002, Laenen, 2009).

In the following sections, we provide illustrations of how we addressed issues in qualitative research with youth (considerations of legal status and protection, developmental capacities, and power) in terms of access and collaboration with gatekeepers, confidentiality and the processes of informed assent and consent, and compensation.

Access to youth and collaboration with gatekeepers

Gaining access to youth for qualitative research touches on issues of their legal status and the protective role played by adults. Whereas most potential adult participants are legally free agents, involving youth participants requires intermediaries to provide access and legal consent (e.g. parents, guardians, teachers, program staff, and government social workers). Relative to research with adults, research with children requires communication and collaboration with more people, and a systematic process of developing good relationships (Cree et al., 2002). As more gatekeepers are involved in accessing youth participants, researchers need to vet and address greater challenges presented. For example, while some agency gatekeepers readily see merit in a study and promote it to colleagues and parents, others may be skeptical about its value or be overburdened with competing job pressures.

In our studies of Mexican American adolescents’ experiences of dating violence (Lela) and of middle school girls’ views on aggression among girls (Joan), we found it logistically easier to draw our sample from a place where the population was already convening and in a space where adolescents would feel comfortable talking to researchers. This means working with schools or with agencies that work with special populations (e.g. shelters for homeless youth, domestic violence shelters, LGBT community centers). Once we identified potential school or agency collaborators, we considered potential benefits for their involvement and planned accordingly. Unlike study participants, schools and agencies are rarely compensated with monetary or other tangible incentives, even though they often invest significant time and resources into the research (e.g. talking to youth about the project, gathering parent consent forms, making available a room to conduct interviews or focus groups). Thus, it is important to identify how the data collection will help the organization achieve its own goals. For example, Tally and Lela provided verbal and written data reports to agencies that were used in staff trainings and grant applications. In some instances, it is often feasible to include questions in the research that elicit data for the purpose of service evaluation in addition to other areas of inquiry more central to the aims of the project. This was particularly true for Susan’s research with LGBT youth centers; wherein the research helped establish funding need for more staff and programs.

Typically, schools and agencies have a formal application process through which the researcher articulates project goals, expectations of staff to successfully conduct the study, and benefits to stakeholders. This process provides opportunities to solicit agency input about how they might use the data, or additional data that could be collected (as long as it is not discordant with research aims). Social workers are in an optimal position to offer services to schools and agencies that are often lacking in resources (e.g. workshops to children and/or parents on research topic area, training sessions for counselors or staff).

Once buy-in is obtained from each school or agency, it is critical to identify key individuals who are enthusiastic about the project and have access to youth. School and agency personnel play an important role as they lend credibility to the project, communicate the importance of participating to both youth and parents, and are often the ones reminding youth to return signed assent or parent consent forms. They also establish trust with parents, as they are often a primary point of contact. Key personnel who assist with the pragmatic tasks associated with the research such as recruitment, obtaining consent, and securing necessary time and physical space may not be those with the highest position of authority, but those who interact daily with youth (e.g. front office personnel, program staff). Without the support of the schools’ and agencies’ decision makers and key personnel, gaining access to children and collecting data can make a qualitative research project more difficult or even unfeasible.

Collaborative relationships with the key personnel were essential for all of our respective research projects. In the case of Joan’s focus group research in school settings, to succeed in gaining access and securing participation it was necessary to consider the school’s focus on educational achievement and accountability and the administrators’ reluctance to release students from academic classes during the school day. Scheduling the focus groups required working with administrators to find the creative solution of meeting for several shorter periods after school when youth were provided access to the “late” bus transportation.

Engaging the support of key personnel with a shared interest in the research topic (e.g. girl fighting) and knowledge of school structure and resources (e.g. access to a private room away from busy corridors) facilitates planning focus group sessions (location, time, duration), and attendance (by encouraging submission of parental consent forms, reminding participants about group meetings, and bringing participants to research groups). Such collaboration with staff can facilitate trust building and ease the entry into the focus group for participants who may feel apprehension about discussing an emotionally laden topic with a stranger.

Ongoing communication with key personnel shows caring and appreciation for their roles, consideration of the multiple tasks they perform and genuine interest in the school or agency and students beyond the research topic, all aiding the collaborative relationship. When problems arise, that impact the research, decisions can be made based upon mutual respect and understanding. In Joan’s middle school groups, the assistant principal did not support groups composed of single ethnicities (Caucasian, Latina or African-American) because she favored girls communicating across groups. Appreciation of her position resulted in her continued wholehearted support of the groups and Joan was able to facilitate the communication among diverse ethnic groups.

Consent, assent and confidentiality

As noted, adults’ traditional role of protecting children raises considerations of finding the balance between children’s needs for protection from harm and their rights to participate, act autonomously, and benefit from research (Fisher et al., 2013; Kirk, 2007; Mishna et al., 2004). Alderson (2005) raises the issue of whether parental consent is always needed if children want to participate, but their parents refuse consent. Mishna et al. (2004) note the absence of literature discussing this as well as the opposite situation: What if parents consent, but children do not assent? Children’s tendency to regard adults as authority figures, and their relative lack of experience in being assertive (Duncan et al., 2009) render ethical concerns related to the imbalance in the researcher–participant power relationship even more troublesome relative to research with adults (Morrow and Richards, 1996). Many of us conduct our qualitative research with vulnerable youth populations (e.g. oppressed youth, youth with disabilities), necessitating extra preparation and careful consideration of potential ethical concerns.

Consent/Assent

Challenges relating to consent and assent arise in all research with youth, but are prominent when recruiting youth in acute distress or crisis. For example, Tally has noted that parents/guardians agree to their children’s participation to a qualitative research study on mental health crises, while appearing harried and distracted. This raises concerns about the extent to which potential participants are weighing the risks and benefits associated with providing descriptive, often intimate, personal information, and worry that the consent and assent provided are not meaningful. Related is the extent to which the assent processes are adequately tailored to the range of developmental levels of child participants so that each child understands their rights and the nature of the study (Alderson, 2005; Kirk, 2007; Mishna et al., 2004). Not only do children need the freedom to refuse participation, or express views or withdraw freely (Kirk, 2007), but they also need concrete and specific information on how to refuse or withdraw, and periodic check-ins on their willingness to continue throughout data collection (Kirk, 2007; Mishna et al., 2004).

In response to these issues, we have found it important to slow down the process, read information aloud, repeatedly ask if there are any questions, and make a concerted effort to be sensitive to youths’ subtle, non-verbal expressions of discomfort or need for clarification during the assent process. We consider that youth have less life experience to draw on for anticipating the nature of questions they will be asked and their possible reactions to sharing detailed, personal information in the context of a prolonged interview (David et al., 2005). Hence, in the processes of obtaining informed consent we provide youth and parents written examples of questions that will be asked about mental illness stigma and the social effects of mental illness, so that they have a vivid sense of what the interviews will entail. For assent forms and other written study materials, as recommended by various researchers (Cree et al., 2002; Lambert and Glacken, 2011), we provide “child-friendly” materials (vetted by same-age non-study individuals) conducive to the comprehension by children with a wide variety of reading and comprehension competencies. Respectful and effective communication includes carefully listening to the child’s communication style and using language the child would use (Eder and Fingerson, 2002). For very young children, oral assent may be more appropriate than asking them to write their name on a document, and pictures can be used creatively in assent documents (Lambert and Glacken, 2011; Yazejian and Goldman, 2013).

Scholars note that the relational aspect of qualitative research might increase the likelihood of a child disclosing a risky behavior or experience (Duncan et al., 2009), or might increase the potential for embarrassment related to personal disclosure (Rich and Ginsburg, 1999). Balancing the aforementioned developmental, power, and legal context of research with children, the assent process should facilitate children’s understanding of all risks, including a potential breach of confidentiality due to child abuse or neglect reporting, and clarity on how any information they share would be conveyed to adults (Duncan et al., 2009; Kirk, 2007; Mishna et al., 2004).

Like other qualitative researchers who include particularly vulnerable children in their studies (Cashmore, 2006; Gorin et al., 2008), we have found it important to extend the assent process with them past the initial formal procedure and to ‘check in' and verify assent at different times throughout the interview (e.g. “Are you still feeling OK?”, “Sure you would like to continue?” or “Would you like to take another break?”). Participants are reminded that they may skip a question or withdraw from the study altogether; and we give them concrete examples of how to communicate this (e.g. “just say ‘next please’”). We use these methods in recognition of underlying developmental and power considerations as well as the potential for the relational nature of qualitative approaches to increase participant discomfort with asking for help or clarification, expressing concerns, or asking to withdraw (Duncan et al., 2009; Mishna et al., 2004).

Such modifications are not meant to be paternalistic or to question youth participants’ coping capacity. Instead, they are geared toward addressing the developmental gap in experience and confidence necessary to assert self-interest in an unfamiliar research situation, a concern that is particularly warranted with youth diagnosed with mental health conditions that often have a negative self-image. Invitations to re-visit assent throughout the research process communicates to children that the goals of the research do not supersede their welfare. At the same time, it is also important to avoid setting up an expectation that the research should or would be too painful for them to tolerate, or to send the message that the researcher under-estimates their strength and capacity in any way. This calls for striking a balanced tone somewhere between concern and positive expectancy.

Context for waiver of parental consent

Some circumstances warrant researchers to move beyond slight modifications of procedures and advocate for major shifts to ensure voices of vulnerable child populations do not remain silent. Family and social dynamics of LGBTQ youth, the focus population of Susan’s work, necessitate modifying recruitment and consent procedures. A pressing barrier to gaining information on this population is the inability to reach vast numbers who are not “out” to family or community, because of factors leading many to keep their sexual and gender minority identities secret, such as family belief systems, regional political climate, and repercussions of coming-out to family, community, and peers (D'Augelli et al., 1998, 2006; Davis et al., 2009; Saltzburg, 2005). Although some attend LGBTQ youth groups or programs with parents’ knowledge, introducing no additional barriers to obtaining parental consent for potential study participation, many others seek out youth programs without informing guardians. In this context, regulation requirements for parent/guardian consent lead to systematic exclusion of LGBTQ youth, thereby further obscuring and marginalizing their lives.

Using Elze’s (2008) work citing the US “federal waiver for parental consent” (45CFR46.408(c)) to establish the rationale for the assent process with LGBTQ youth, Susan’s advocated with the university’s Institutional Review Board (IRB) for a waiver of parental consent. Such action is warranted in cases in which parental consent may result in harm of the child or infringe on their rights to privacy, or unjust exclusion. Citing the literature, she argued that the family crisis that youth disclosure may create when asking parents for consent (including rejection, relationship distancing, and abuse) justifies the consent waiver. As part of her argument, she invoked the ethical responsibility of social work researchers to find alternate ways to include these children in their research. Following recommendations set forth by predecessors conducting mental health research with children (Fisher, 1993; Fisher et al., 1996) and research with LGBTQ youth (D'Augelli and Hershberger, 1993; Elze, 2008), she successfully implemented the assent process employing “youth research advocates” in place of parent/guardian consent. These advocates were licensed social workers who met both individually and collectively with the youth before and after the interviews to ensure youth understanding, assess emotional vulnerability, process the experience, and provide further follow-up if needed. The youth research advocacy team provided an essential safety-net for youth participants in keeping with ethical considerations.

Confidentiality

Concerns about nonmaleficence (avoidance of harm), especially pertaining to disclosure of abuse or neglect, are weighed against minors' right to confidentiality, on a case-by-case basis. There are no clear professional guidelines provided by the National Association of Social Workers (2008) Code of Ethics (or other professional codes) for when to prioritize one ethical imperative over the others in the context of research or practice. Hence, researchers are left to develop heuristics for situations that warrant breach of confidentiality for the sake of child protection.

In qualitative research, quandaries about upholding confidentiality regarding vs. protectiveness may occur more frequently relative to other types of research. Given the opportunity to talk about relationships, mental illness or sexual and gender identities in ways that they may not have been able to do so previously, and by virtue of the highly personal information that children share with interviewers, qualitative researchers often “step into” a relationship with participants. In the case of vulnerable youth, such as LGBTQ youth who are hungry for adult approval, adult role models, and relationships with adults who have walked in their shoes, this perception of a relationship becomes heightened when the researcher is transparent about her or his own sexual or gender minority identity. The interest, empathy, and connection displayed by the researcher, in the context of free-flowing nature of interactions, may be understood by youth as signs of an emerging personal or therapeutic relationship. Amy, like Susan and others, found that the warm interpersonal environment, often recommended in research with young people (Eder and Fingerson, 2002; Kirk, 2007; Mason and Hood, 2011), promotes participants’ sense of a reciprocal relationship. In this context, qualitative research often yields, or even calls for, personal revelations of hurt or distress more than other types of research, some of which may be unknown to significant others, and which may raise concerns about current or potential risk of harm to child participants (Cashmore, 2006; Duncan et al., 2009; Kirk, 2007; Lambert and Glacken, 2011; Mishna et al., 2004). What then may seem to participants like a thinning power differential between the participant and researcher, may be misleading in a sense because the more vulnerability and trust demonstrated by the participant, the greater the fiduciary obligation on the part of the researcher to protect her participant from harm and ensure that the research relationship is safe (Kutchins, 1991).

In response to such revelations, as mandated reporters, researchers are morally and legally obligated to breach participant confidentiality and report the risk to another party that can intervene on the child’s behalf. Hence, in any study designed as “confidential”, researchers must clearly communicate to minors and their parents or other gatekeepers prior to study engagement that complete confidentiality cannot be guaranteed throughout the research process and to outline explicitly those situations that warrant breaking it.

To minimize instances of breach of confidentiality, Tally’s research team empowered youth to have maximal control over information by informing in advance what would prompt exploration and possible disclosure: disclosed or suspected abuse, neglect, self-harm, or indication that the participant poses danger to others. They are also reminded that they have full control over what they disclose and always have the option of avoiding answering any questions or ending their participation in the research at any time. In this regard, as noted above, we strive to draw distinctions between the researcher role on the one hand, and the role of therapist, advocate, mentor, or friend on the other hand, so as to provide the child sufficient information to make a decision about what to disclose.

Inevitably, however, dilemmas related to confidentiality will arise in the context of conducting qualitative research with children. These dilemmas often center on whether to insist on discussion of the disclosed information with guardians or other authorities, and whether to follow-up with the child/family after their participation in the research. The competing consideration is often the desire to protect the minor's rights to confidentiality and family privacy, a desire to avoid unnecessary stress for the child and/or family, and a desire to protect the child’s positive experience as a research participant (MacDonald and Greggans, 2008).

In Tally’s research with youth with diagnosed mental disorders, not infrequently, participants are involved in various social services and have a known history of abuse, neglect, or self-harm. We do not obtain releases for information but typically rely on youth's self-report that the reported risk (e.g. self-harm or abuse) is known to significant others. In such cases, questions can sometimes arise about the extent to which a participant is truthful about previous disclosure. Or even when it is evident that others are aware of some of the risk reported, the researchers may be uncertain about how fully others are aware of the breadth, intensity or recency of the issues disclosed during the research interview. When situations of ambiguous risk arise, interviewers are instructed to halt the interview and engage in a series of steps of information gathering. The interviewer reminds the participant that concerns about their safety may necessitate breaking confidentiality. Subsequently, specific questions to help assess the nature and likelihood of previous disclosure and intervention can be posed (and responses documented): Who knows? When and how did they find out? What did they do about this? What were the outcomes of what was done? If the answers to these questions are (perceived) specific and credible, the research interview may proceed, and debriefing with the principal investigator can occur after the interview. If answers are non-specific or perceived as unrealistic, however, the interview can be halted and the principal investigator contacted right away.

This course of action, when it occurs, essentially privileges the youth’s right to be protected from harm over the participant’s unguaranteed right to confidentiality because, as social work researchers, we regard the researcher–participant interaction as a fiduciary relationship (much like between client and social worker). Herein, the fiduciary has “(1) special duties because of the trust or confidence reposed in the fiduciary; (2) the fiduciary has special powers…because of the nature of the relationship; and (3) as a consequence, the fiduciary must act in the best interests of the client (participant) and cannot take advantage…” (Kutchins, 1991: 107).

These specific guiding processes and principles (beyond delineating circumstances warranting mandated reporting) should be incorporated in the training of all individuals conducting the research, and elaborated in IRB applications. These not only help ensure that disclosed risk does not go unheeded, but also can also help minimize researchers’ stress.

As it is not possible to anticipate every situation that arises, an important principle is to develop supportive and accessible relationships within the research team. For example, encouraging post-interview debriefing in cases of any doubt or concern; research assistants can be encouraged to contact the principal investigator during or directly following an interview or focus group (if they can find some privacy). A collaborative research team approach that provides opportunities for processing questions and concerns related to specific interviews helps with ethical decision-making and reducing researcher strain in the context of research with children (Gorin et al., 2008). In addition, it may be helpful to forge links with experts in child abuse and protection and mental health that the researcher can consult with about cases that raise concerns.

Compensation

Our literature review of research practices with child and youth participants found little discussion of compensation. Methods of compensation touch on developmental considerations and power imbalances, as adult researchers seek to recognize the contributions of child participants in ways that are meaningful but not potentially coercive. As with adults, various options are available (e.g., store gift cards, cash, movie passes).

One challenge of compensation for children is that an amount of financial compensation that may not be a significant incentive with adults, may border on coercion with children and youth. In Lisa’s study of youth aging out of the child welfare system, compensation for 90 min interviews was $15. Several participants indicated that their initial motivation to participate in the study was because they desperately needed money. The manner in which compensation is provided to participants can reduce power differentials between them and the researcher (David et al., 2005; Eder and Fingerson, 2002). To ensure participants did not feel coerced to complete the interview, youth participants were given the $15 at the time of assent/consent and told they could leave at any time. Comments and jokes about leaving without answering any questions often followed, yet no one chose to do so.

In Amy’s study with mothers and their children (aged 12–14) in domestic violence shelters, issues of compensation were approached considering both material compensation and interpersonal processes that might be rewarding to participants, with attention to power differentials in the forefront. Material compensation consisted of a $20 department store gift card for the family, given to the mother, and a $5 coupon for a popular fast food restaurant given to each youth participant. These forms of compensation were determined in collaboration with shelter personnel. The likelihood that each family included siblings who did not meet study inclusion criteria was taken into consideration; this observation influenced the decision, as a matter of justice, to have a larger compensation amount for the mother that could benefit the whole family. Every youth had a choice between two restaurants for her or his coupon and every mother had a choice between two stores (i.e. a grocery store or a large department store offering clothing, gas, etc.). As in Lisa’s child welfare study, participants were compensated prior to their interviews, immediately after informed consent/assent procedures, to reinforce that they could stop participation at any time or pass on any question and all participated fully. Secondly, youth left their interviews with a colorful life story timeline which they co-created as part of data collection to portray their important experiences and relationships. After each interview, the timeline was photocopied and the participant kept the original. This contributed to a sense of reciprocity, and emphasized that the story belonged to the youth.

Social work and qualitative research with youth: A good fit

Social work researchers have a unique capacity to conduct qualitative research with youth and to negotiate the logistical challenges and ethical considerations involved. Social workers are trained and skilled in ethical decision-making, collaborating with various stakeholders, monitoring boundaries in interpersonal communications, and assessing as well as reporting high-risk situations. Moreover, knowledge of human development and understanding of power differentials are foundational in social work education. Ethical codes, central in social work, explicitly outline values such as honoring every individual’s worth and dignity along with calls to serve the vulnerable and disenfranchised (for example, see British Association of Social Workers, 2012; National Association of Social Workers, 2008). With such an educational background, training, and value orientation, the development of research protocols to address ethical dilemmas should be second-nature. Social workers maintain connections to the community, agencies and institutions where children are, and social workers’ interpersonal skills are assets for nurturing such relationships.

The experiences of six social work researchers conducting qualitative studies with diverse youth populations, described in this article, included successful approaches to access, consent, assent, confidentiality, and compensation. Our aim is to contribute to the literature practical solutions to such logistical challenges, however, primarily, we share our perspectives in the hopes of encouraging others to conduct qualitative social work research with youth. It has the potential to increase our understanding of their lives and perspectives, and to honor their right to have a voice in matters that concern them.

References

Alderson

(2005) Designing ethical research with children. In: Farrell

(ed.) Ethical Research with Children, New York, NY: Open University Press.

British Association of Social Workers (2012) The code of ethics for social work: Statement of principles, Birmingham: The British Association of Social Workers.

Cashmore

(2006) Ethical issues concerning consent in obtaining children's reports on their experience of violence. Child Abuse & Neglect 30: 969–977.

Chanmugam

(2014) Got one another's backs: Mother-teen relationships in families escaping domestic violence. Journal of Human Behavior in the Social Environment 24: 811–827.

Checkoway

(2011) What is youth participation? Children and Youth Services Review 33: 340–345.

Coyne

(2010) Research with children and young people: The issue of parental (proxy) consent. Children & Society 24: 227–237.

Cree

Kay

Tisdall

(2002) Research with children: Sharing the dilemmas. Child and Family Social Work 7: 47–56.

D'augelli

Hershberger

(1993) Lesbian, gay and bisexual youth in community settings: Personal challenges and mental health problems. American Journal of Community Psychology 21: 421–448.

D'augelli

Hershberger

Pilkington

(1998) Lesbian, gay, and bisexual youth and their families: Disclosure of sexual orientation and its consequences. American Journal of Orthopsychiatry 68: 361–371.

10.

D'augelli

Grossman

Starks

(2006) Childhood gender-atypicality, victimization and PTSD among lesbian, gay and bisexual youth. Journal of Interpersonal Violence 21: 1462–1482.

11.

David

Tonkin

Powell

(2005) Ethical aspects of power in research with children. In: Farrell

(ed.) Ethical Research with Children, New York, NY: Open University Press.

12.

Davis

Saltzburg

Locke

(2009) Supporting the emotional and psychological well being of sexual minority youth: Youth ideas for action. Children and Youth Services Review 31: 1030–1041.

13.

Duncan

Drew

Hodgson

(2009) Is my mum going to hear this? Methodological and ethical challenges in qualitative health research with young people. Social Science & Medicine 69: 1691–1699.

14.

Eder

Fingerson

(2002) ) Interviewing children and adolescents. In: Gubrium

Holstein

(eds) Handbook of Interview Research: Context and method, Thousand Oaks, CA: SAGE.

15.

Elze

(2008) 8,000 miles and still counting… Reaching gay, lesbian, and bisexual adolescents for research. In: Meezan

Martin

(eds) Research Methods with Gay, Lesbian, Bisexual and Transgender Populations, New York, NY: Harrington Park Press.

16.

Farrell

(2005) Ethical Research with Children, New York, NY: Open University Press.

17.

Fisher

(1993) Integrating science and ethics in research with high-risk children and youth. Social Policy Report: Society for Research in Child Development 7: 1–27.

18.

Fisher

Brunnquell

Hughes

(2013) Preserving and enhancing the responsible conduct of research involving children and youth: A response to proposed changes in federal regulations. Social Policy Report 27: 1–23.

19.

Fisher

Hoagwood

Jensen

(1996) Casebook in ethical issues in research. In: Hoagwood

Jensen

Fisher

(eds) Ethical Issues in Mental Health Research, Mahwah, NJ: Lawrence Erlbaum Associates.

20.

Fleming

(2010) Young people's involvement in research: Still a long way to go? Qualitative Social Work 10: 207–223.

21.

Flores KS (2008) Youth participatory evaluation: Strategies for engaging young people. San Francisco, CA: John Wiley & Sons, Inc.

22.

Gorin

Hooper

Dyson

(2008) Ethical challenges in conducting research with hard to reach families. Child Abuse Review 17: 275–287.

23.

Hart

(2008) Stepping back from “The Ladder”: Reflections on a model of participatory work with children. In: Reid

AEA

(ed.) Participation and Learning, NY: The City University of New York.

24.

Head

(2011) Why not ask them? Mapping and promoting youth participation. Children and Youth Services Review 33: 541–547.

25.

Hester CJ (2004) Adolescent consent: Choosing the right path. Issues in Comprehensive Pediatric Nursing 27: 27–37.

26.

Kanner

Langerman

Grey

(2004) Ethical considerations for a child's participation in research. Journal for Specialists in Pediatric Nursing 9: 15–23.

27.

Kirk

(2007) Methodological and ethical issues in conducting qualitative research with children and young people: A literature review. International Journal of Nursing Studies 44: 1250–1260.

28.

Kutchins

(1991) The fiduciaty relationship: The legal basis for social workers' responsibilities to clients. Social Work 36: 106–113.

29.

Laenen

(2009) I don't trust you, you are going to tell', adolescents with emotional and behavioral disorders participating in qualitative research. Child: Care, Health and Development 35: 323–329.

30.

Lambert

Glacken

(2011) Engaging with children in research: Theoretical and practical implications of negotiating informed consent/assent. Nursing Ethics 18: 781–801.

31.

Letendre

Smith

(2011) “It's murder out today”: Middle school girls speak out about girl fighting. Children & Schools 33: 47–57.

32.

MacDonald

Greggans

(2008) Dealing with chaos and complexity: The reality of interviewing children and families in their own homes. Journal of Clinical Nursing 17: 3123–3130.

33.

MacNaughton

Smith

(2005) Transforming research ethics: The choices and challenges of researching with children. In: Farrell

(ed.) Ethical Research with Children, New York, NY: Open University Press.

34.

Mason

Hood

(2011) Exploring issues of children as actors in social research. Children and Youth Services Review 33: 490–495.

35.

Mishna

Antle

Regehr

(2004) Tapping the perspectives of children: Emerging ethical issues in qualitative research. Qualitative Social Work 3: 449–468.

36.

Morrow

Richards

(1996) The ethics of social research with children: An overview. Children & Society 10: 90–105.

37.

Moses

(2009) Self-labeling and its effects among adolescents diagnosed with mental disorders. Social Science & Medicine 68: 570–578.

38.

Moses

(2011) Stigma apprehension among adolescents discharged from brief psychiatric hospitalization. Journal of Nervous & Mental Disease 199: 778–789.

39.

Mullender A, Hague G, Imam U, et al. (2002) Children's perspectives on domestic violence. London: SAGE.

40.

National Association of Social Workers (2008) Code of ethics of the National Association of Social Workers, Washington, DC: National Association of Social Workers Press.

41.

Rich

Ginsburg

(1999) The reason and rhyme of qualitative research: Why, when, and how to use qualitative methods in the study of adolescent health. Journal of Adolescent Health 25: 371–378.

42.

Saltzburg

(2005) Co-constructing adolescence for GLBT youth and their families. In: Mallon

McCourt-Hess

(eds) Child Welfare for the Twenty-First Century: A Handbook of Practices, Policies, and Programs, New York, NY: Columbia University Press.

43.

Saltzburg

Davis

(2010) Co-authoring gender-queer youth identities: Discursive tellings and retellings. Journal of Ethnic & Cultural Diversity in Social Work 19: 87–108.

44.

Schelbe L (2013) “Some type of way”: An ethnography of youth aging out of the child welfare system. PhD Thesis: University of Pittsburgh, PA.

45.

Staller KM (2008) Children's rights. In: Mizrahi T and Davis L (eds) Encyclopedia of Social Work. 20th ed. Washington, DC: National Association of Social Work Press and Oxford.

46.

Ungar

Joffe

Kodish

(2006) Children are not small adults: Documentation of assent for research involving children. The Journal of pediatrics 149: S31–S33.

47.

Williams

(2014) Experiences with violence in Mexican American and European American high school dating relationships. Children & Schools 36: 115–124.

48.

Yazejian

Goldman

(2013) Research with young children. Social Policy Report 27: 18–19.

49.

Zeldin

Camino

Mook

(2005) The adoption of innovation in youth organizations: Creating the conditions for youth-adult relationships. Journal of Community Psychology 33: 121–135.

50.

Zeldin

Petrokubi

Macneil

(2008) Youth-adult partnerships in decision making: Disseminating and implementing an innovative idea into established organizations and communities. American Journal of Community Psychology 41: 262–277.