Chapter 31: Ethical considerations in conducting research with hard-to-reach populations

Tatiana Corrales

Learning outcomes

Upon completion of this chapter, you should be able to:

  • Understand some of the factors that make populations hard to reach.
  • Understand some of the ethical challenges associated with research involving hard-to-reach populations.
  • Develop study protocols to mitigate some of the ethical challenges and to empower ‘hard-to-reach’ populations to participate in research.

Who and what are hardtoreach populations?

The term ‘hard to reach’ is not well defined, but in general is used to refer to groups or populations that are difficult to engage in research.1 The factors that make a population hard to reach are varied and may reflect institutional barriers (e.g. gatekeeping by service providers or government bodies) and/or individual characteristics of people within that population (e.g. homelessness, substance abuse, age, the presence of disabilities, or people who highly dependent on medical care).1,2,3 Due to their hidden nature, these populations are often excluded from research. This chapter focuses on the factors that can make certain populations hard to reach and the ethical implications of not including vulnerable groups in research.

‘Hard-to-reach’ populations are typically considered vulnerable, marginalised and/or disenfranchised. Normally, hard-to-reach populations are those that exist outside of the mainstream, such as people who are homeless, children in Out-of-Home Care (OOHC), people in prison and people experiencing multiple forms of disadvantage or hardship.3 Hard-to-reach populations can be characterised by significant histories of trauma. Their interactions with systems of care and control have often exacerbated their trauma, leading to increased wariness and suspicion of systems, services and institutions, including universities.3,4,5

There are also certain personal characteristics, including age (specifically children) and the presence of intellectual disabilities or cognitive impairments that can lead to groups being considered vulnerable and, consequently, hard to reach.4,6,7,8

Contested terminology

While terms like ‘vulnerable’, ‘marginalised’ and ‘disadvantaged’ are commonly used to describe hard-to-reach populations, these terms are often ill-defined and contested.3,6 Vulnerability is not a static construct: its definition depends on context and perspective. From a research governance perspective (i.e. research ethics), vulnerability is equated with risk – risk to the vulnerable individual and risk to the organisation.7,9 This focus on managing risk creates its own set of challenges, including significant barriers that ultimately exclude certain cohorts of people from research.

However, vulnerability – particularly when it is grounded in trauma – can coexist with risks that require management or control. For example, children and young people who have experienced maltreatment are considered highly vulnerable. These same children may exhibit challenging behaviours, including antisocial and criminal activity.10 Vulnerability can therefore elicit compassion and protection, but only when it conforms to certain stereotypes of what it means to be vulnerable.3

The attributions applied to vulnerable people may not reflect how people define themselves. The term ‘vulnerability’, for example, can connote weakness and frailty, or the need for protection. It can also be used to imply that vulnerable people lack the capacity for self-protection and personal agency.4,6,11 Similarly, the term ‘marginalised’ has connotations of personal responsibility or personal failure.12 However, marginalisation is a process of exclusion from the mainstream, driven by entrenched systemic and structural inequities.13

The language we use to describe research participants is important and has implications for participants’ willingness to engage with our research. It may not be that populations are ‘hard to reach’ but that our way of thinking about and doing research needs to be more flexible, creative and inclusive.8

Table 31.1 provides a non-exhaustive list of groups that are typically considered vulnerable and consequently hard to reach. It is important to note that many of the characteristics in the table are not mutually exclusive. In fact, it is the intersection of multiple individual characteristics with institutional barriers that render certain populations hidden and silenced.

Table 31.1: Hard-to-reach populations and factors that increase perceived risks of research participation

Population Perceived risks and vulnerabilities
Children (0–18 years old) Cognitive immaturity

Legally unable to provide free and informed consent (up to certain age thresholds)

Potential for distress and/or traumatisation perceived as unacceptably high
Children in statutory Out-of-Home Care (OOHC; includes foster, kinship, and residential care) Under the guardianship of the state – participation is contingent on approval from child protection services

Significant trauma histories place this group at an increased risk of distress and/or re-traumatisation

Presence of mental health challenges, intellectual disabilities, neurodivergence (e.g. Autism Spectrum Disorder, Attention-Deficit/ Hyperactivity Disorder) and/or behavioural difficulties14
Children under the supervision of youth justice systems Similar characteristics to children in OOHC

Require approval from youth justice departments

Where children are detained, approval from the detention centre is also required
People experiencing homelessness High risk of distress and re-traumatisation

Presence of mental health difficulties and/or substance abuse histories

Perceived unsafe living environments
People experiencing significant mental health difficulties High risk of distress and re-traumatisation

Capacity to provide free and informed consent may be compromised
People living with disabilities (intellectual and/or physical) For people with intellectual disabilities, capacity to provide free and informed consent may be limited or restricted. May be dependent on others to assist with communication. Raises concerns of coercion.
People with dementia As above
People from culturally and linguistically diverse backgrounds (including recently arrived immigrants and refugees) Capacity to provide free and informed consent may be influenced by challenges with English proficiency. The use of interpreters can raise concerns about judgement and stigma within a community, particularly if the research is touching on sensitive topics (e.g. child protection intervention, reproductive health, family violence)15,16
First Nations People High risk of distress and re-traumatisation through culturally unsafe and inappropriate research methodologies

Risks about how data will be interpreted and disseminated – with/without community consultation? Risk of inappropriate interpretations of data if community not involved in collection and analysis of data.
People with substance abuse challenges Safety concerns (for participants and researchers)

Capacity to provide free and informed consent may be compromised
Victim-survivors (child abuse, domestic and family violence) High risk of distress and re-traumatisation

Participant safety a key concern – need to ensure that participation in the study does not increase the risk of violence by intimate partners

People in custodial settings (prisons, immigration detention) High risk of distress and re-traumatisation

Safety (of participants and the researcher)

Real of perceived coercion to participate in research
People who are highly dependent on medical care and/or are hospitalised Risk of real or perceived coercion due to fear that medical treatment will be withheld

Inability to walk away from the hospital setting (for hospitalised patients)
People who are pregnant Perceived risks vary by nature of the study: Medical research – perceived risk of procedures that may harm the foetus (not the necessarily the mother)

Social science research – perceived risk of causing distress.

As the information in Table 31.1 shows, from a research governance perspective vulnerability is framed through a lens of risk: that participants will experience distress and trauma, that the safety of participants and researchers is at risk and that there is potential for unethical conduct if the correct protocols are not in place to ensure participants are able to provide free and informed consent.

Ensuring that risks are properly managed is an important element of all research. However, when vulnerability is so narrowly equated with risk, people’s agency – their right to determine whether they want to participate in research – is eroded. In attempting to protect vulnerable participants from unethical practices, we run the risk of excluding their voices from research, policy and practice.3,7

Institutional barriers

A group’s perceived vulnerability can also be used by institutions – including research ethics committees, service providers and statutory bodies (e.g. child protection, criminal legal systems) – to limit access to participants. This is commonly referred to as ‘gatekeeping’.3,7,9

Often it is important to rely on the discretion of people who have a deep knowledge and understanding of the group that is being approached to participate in research. As researchers, we may not always understand the complexity of the lives of participants. Or we may be unprepared for the ways in which participants respond to the research. ‘Gatekeepers’ can provide important information to ensure that participants are approached with sensitivity.

However, gatekeeping can also be a form of control, grounded in paternalistic ideals. This is particularly evident with populations that are deemed simultaneously vulnerable and risky or dangerous. Box 31.1 provides an example of gatekeeping by statutory bodies on a project exploring the service needs and experiences of mothers in prison.

Box 31.1: Barriers to recruiting criminalised women who have been pregnant and/or given birth in prison

This project involved in-depth interviews and focus groups with a range of stakeholders, including prison staff, lawyers, service providers and women with lived experience. Initially, ethics approval was sought to interview women in custody. Access to prisons was limited during multiple extended lockdowns during the COVID-19 pandemic. Further, the corrections department’s policies limited how criminalised women – including those on community-based orders or who had been released on parole – could participate in the study. For example, they could not be reimbursed for their time participating in the study. Women in custody could be invited to participate in the study, but their involvement had to be facilitated through the prison, and interviews had to occur online. This raised ethical, methodological and logistical concerns. For example, recruitment would need to be undertaken by prison staff as the researcher was not allowed on site. This meant that there was no way to guard against selection bias, or to minimise the potential for perceived coercion. It was also impossible to guarantee women’s privacy during online interviews conducted within a custodial setting. This was especially concerning given the focus of the study on women’s experience of the ‘care’ they had received from the prison system during pregnancy and following childbirth. As none of these factors could be guaranteed, and as the prospect of future extended lockdowns remained the researchers decided to recruit women in the community, rather than attempting to access women in custody. The result was that the study did not capture the experiences of women who were in custody during the period of the study.

Based on an unpublished research project undertaken by Tatiana Corrales (2021–2023)

Research involving vulnerable populations sometimes entails balancing competing ethical and practical considerations. From a practical perspective, there are often external constraints linked to funding and reporting timelines, that limit the amount of time that can be spent on recruitment activities. In these situations, it is sometimes necessary to compromise certain elements of the study – in this case, giving voice to women in custody – in order to safeguard the rights of vulnerable participants.

Gatekeeping may not represent an attempt to prevent certain groups of people from participating in research. The reasons for gatekeeping are varied and in general are intended to protect vulnerable groups from exploitation.9 However, whether intentionally or unintentionally, gatekeeping can contribute to a population being hard to reach’ This, in turn, can result in certain groups being excluded from research, thereby diminishing their voices and making them increasingly hard to reach, in an endless loop.

The ethics of research with hardtoreach populations

In Australia, the ethical conduct of human research is grounded in the principles of merit and integrity, justice, beneficence and respect.17

There is a tension inherent in seeking to balance these principles. Ensuring that participants are not exposed to, or directly caused, harm as a result of research is fundamentally important. However, participants also have the right to decide whether they want to participate in research. This is embodied in the principle of respect, which states that ‘respect for human beings involves giving due scope, throughout the research process, to the capacity of human beings to make their own decisions.’17(p11) As researchers, we have an ethical obligation to support vulnerable groups to be active participants in research.

Ethical research involves more than just managing risks. It is about genuine engagement with communities and research participants, based on respect for their autonomy, dignity and expertise in their own lives. The inclusion of people with lived experience in research project – from conceptualisation to dissemination of findings – can be a powerful means of empowering vulnerable populations to be heard.18 The slogan ‘nothing about us without us’18,19 highlights the importance of genuine engagement, ensuring that people with lived experienced are consulted and included in every stage of research.

Additional ethical considerations with hardtoreach populations

  • As far as practicable, engage people with lived experience early in the research process – preferably before you design the study.
  • Include adequate and appropriate remuneration for lived-experience consultants and participants in the research budget. Consultants should be paid for their time as experts; research participants should be compensated for their time in a way that reflects their contribution. Participants should also be reimbursed for the cost of travel. Paying participants is a contentious area of research ethics. Some research ethics committees are not comfortable with high-value payments to participants, due to the risk that this will act as an inducement to participation, thereby limiting a participant’s capacity to provide free and voluntary consent. What constitutes a ‘high-value’ payment can be arbitrary, but in our research, $80 and $100 payments have been considered excessive by multiple ethics committees. However, high-value payments can be ethically justified as a means of recruiting and retaining research participants, and as a gesture that participants’ views are valued.
  • Develop thorough research protocols that explain how participant distress and participant safety, as well as researcher distress and safety, will be managed. Additional considerations may be needed, depending on the nature of the population. For example, when undertaking research with families involved in the child protection system, it is important to consider:
    • the potential for extreme levels of distress among participants, including threats of suicide
    • legal obligations on the research team to report suspected child maltreatment to relevant authorities, and how this will be managed
    • how the research team will manage participant disclosures of harm perpetrated against them.
  • Ethics should always be about more than managing risk. Ethical research with hard-to-reach populations is fundamentally about valuing the perspectives of people who are often excluded from research. It is about respecting their autonomy to choose whether they want to participate in a study, and about creating safe spaces that will enable their participation.
  • There are unique considerations that apply to research with or about First Nations Peoples. Chapter 19 explores this in more detail.


Undertaking research with hard-to-reach populations can be challenging. Despite these challenges, it is important that researchers attempt to engage with these populations as they have too often been rendered invisible in research and policy discussions. Rather than only seeing their vulnerability and marginalisation as evidence of risk, we should be working to empower the silenced and invisible members of society to contribute to and participate in research. The experiences and perspectives of hard-to-reach populations are valuable in their own right. They also have the power to broaden our understanding and develop more creative and meaningful solutions to entrenched structural problems.


  1. Sydor A. Conducting research into hidden or hard-to-reach populations. Nurse Res. 2013;20(3):33-37. doi: 10.7748/nr2013.
  2. Ellard-Gray A, Jeffrey NK, Choubak M et al. Finding the hidden participant: solutions for recruiting hidden, hard-to-reach, and vulnerable populations. Int J Qual Methods. 2015;14(5):1-10.
  3. Ellis R. What do we mean by a ‘hard-to-reach’ population? Legitimacy versus precarity as barriers to access. Sociol Methods Res; 2022. doi: 10.1177/0049124121995536
  4. Brown K. Vulnerability and Young People: Care and Social Control in Policy and Practice. Policy Press; 2017.
  5. Kendrick A, Steckley L, Lerpiniere J. Ethical issues, research and vulnerability: gaining the views of children and young people in residential care. Child Geogr. 2008;6(1):79-93. doi:10.1080/14733280701791967
  6. Øyvind I, Snipstad M. Concerns regarding the use of the vulnerability concept in research on people with intellectual disability. Br J Learn Disabil. 2022;50(1):107-114. doi: 10.1111/bld.12366
  7. Taplin S, Chalmers J, Brown J, Moore T, Graham A, McArthur M. How do research ethics committee members respond to hypothetical studies with children? Results from the MESSI study. J Empirl Res Hum Res Ethics. 2022;17(3):254-266. doi: 10.1177/15562646221087530
  8. Bradbury-Jones C, Isham L, Taylor J. The complexities and contradictions in participatory research with vulnerable children and young people: a qualitative systematic review. Soc Sci Med. 2018;215:80-91. doi: 10.1016/j.socscimed.2018.08.038
  9. Kay L. Guardians of research: negotiating the strata of gatekeepers in research with vulnerable participants. PRACTICE. 2019;1:37-52. doi: 10.1080/25783858.2019.1589988
  10. Baidawi S, Ball R. Multi‐system factors impacting youth justice involvement of children in residential out‐of‐home care. Child Fam Soc Work; 2022. doi:10.1111/cfs.12940
  11. Samuels GM, Pryce JM. “What doesn’t kill you makes you stronger”: survivalist self-reliance as resilience and risk among young adults aging out of foster care. Child Youth Serv Rev. 2008;30(10):1198-1210. doi: 10.1016/j.childyouth.2008.03.005
  12. Broadhurst K, Mason C. Maternal outcasts: raising the profile of women who are vulnerable to successive, compulsory removals of their children – a plea for preventative action. J Soc Wel Fam Law. 2013;35(3):291-304. doi:10.1080/09649069.2013.805061
  13. Bauman Z. Wasted Lives: Modernity and Its Outcasts. Polity Press; 2004.
  14. Águila-Otero A, Bravo A, Santos I, del Valle JF. Addressing the most damaged adolescents in the child protection system: an analysis of the profiles of young people in therapeutic residential care. Child Youth Serv Rev; 2020. doi:10.1016/j.childyouth.2020.104923
  15. Chase E, Otto L, Belloni M, Lems A, Wernesjö. Methodological innovations, reflections and dilemmas: the hidden sides of research with migrant young people classified as unaccompanied minors. Journal of Ethnic and Migrant Studies. 2020;2:47-473. doi: 10.1080/1369183X.2019.1584705
  16. Berman RC, Tyyskä V. A critical reflection on the use of translators/interpreters in a qualitative cross-language research project. 2011; 10(2): 178-190. doi: 10.1177/160940691101000206
  17. National Health and Medical Research Council. National Statement on Ethical Conduct in Human Research (2007) – Updated 2018. National Health and Medical Research Council; 2018. Accessed June 26, 2023.
  18. Aldridge J. “With us and about us”: participatory methods in research with ‘vulnerable’ or marginalized groups. In: Liamputtong P, ed. Handbook of Research Methods in Health Social Sciences. Springer; 2019:1919-1934. doi: 10.1007/978-981-10-5251-4_126
  19. Charlton JI. Nothing About Us Without Us: Disability Oppression and Empowerment. University of California Press; 1998.