Getting past the gatekeeper: Cultural competence, field access and researching gender-based violence – Evidence from four countries

Background: Increasingly researchers are recognising the importance of including the perspectives of children and young people in research and in the development of interventions and innovations, in order to understand children's lives on their own terms. Purpose: This article reflects on the potential risks posed to children and young people in research and the role of gatekeepers in granting access and, in direct field activities, especially where the research could be deemed as ‘sensitive’. Research Design:Using a combined descriptive and autoethnographic research design, questionnaires and reflections were gathered from the field experiences of researchers in four countries (India, Jamaica, Uganda and the UK) conducting research on children's experiences of violence victimisation as part of a gender-based violence prevention project. Data was collected retrospectively and analysed thematically with the aid of QSR NVivo. Results: Based on the five themes that emerged: accessing schools; approach to access; accessing participants; lessons learnt; and advice for other researchers, the overall findings indicate that researching sensitive topics with children and young people is necessarily problematic in order to safeguard them. This is especially the case in contexts where discrimination and stigma relating to certain topics may be a significant barrier to children's participation. Conclusion: Research planning requires attention to gatekeeper power, safeguarding children and young people, and understanding how processes and attitudes concerning the status and rights of children at national and institutional levels might influence the research process.


Introduction
Since the UN Convention on the Rights of the Child was adopted in 1989 there has been international recognition that children have a right to be heard in matters that affect their lives, and beyond this, they also have a valuable contribution to make. 1 Researching sensitive issues however can be challenging at the best of times, for a range of reasons, and involving children and young people in research on sensitive issues can be especially challenging. Nevertheless, a recognition of the rights of children and young people to be heard, coupled with the changing views of childhood, has led to research which seeks to directly access the views of children and young people rather than accessing their views through proxies as customary in times past (Coyne, 2010). Increasingly researchers are recognising the importance of including the perspectives of children and young people in research and in the development of interventions and innovations, in order to understand their lives on their own terms. That is why the None in Three Research Centre (Ni3) 2 thought it was prudent in designing educational interventions for preventing gender-based violence (GBV) to gather the perspectives of children and young people. As mentioned above, however, researching sensitive issues can be a challenge, exacerbated when children and young people are involved.

Researching sensitive issues
There is growing research interest in sensitive social issues such as sexuality, child abuse and death. Furthermore, with increased public awareness of the impacts of domestic violence, intimate partner abuse and family breakdown, there have also been increased effort to investigate the experiences of people in these situations. The aim has been to gain increased understanding and awareness of the impact of the experience of sensitive issues on people's lives (Cowles, 1988) and to devise prevention strategies (Jones et al., 2017).
What counts as a sensitive research topic or issue is dependent on both context and cultural norms and values. Lee (1993) suggests there are three factors that create a concern about sensitivity. Firstly, whether a topic or issue is considered private, stressful, or sacred (such as sexuality or death). Secondly, whether a topic or issue, if revealed, might cause stigmatisation or fear (such as a HIV+ status). Thirdly, whether a topic or issue relates to the presence of a political threat where researchers may study areas subject to controversy or social conflict. The sensitive nature of the research may not be apparent at the beginning of the research project. Furthermore, a topic or issue presumed to be sensitive in nature may not be. For example although abuse is considered by many in Western societies to be a private and sensitive issue, women, through the #MeToo Movement 3 and elsewhere, report being relieved to be able to talk freely about their experiences (Walker 1979(Walker , 1984McCosker et al., 2001). Additionally, Sieber and Stanley (1988) argue that sensitive studies are those '… in which there are potential consequences or implications, either directly for the participants in the research or for the class of individuals represented by the research' (p.49). Underlining this, Lee (1993) provides that sensitive research '… poses a substantial threat to those who are or have been involved ' (p.4). It should be noted however that these definitions do not restrict the focus to only the participants, nor do they define the nature of risk to all the participants in the research process (Driscoll et al., 1997).
Studying sensitive topics not only presents challenges around gatekeeping but also creates methodological issues for the researcher, including framing the topic/issue, defining and accessing the sample (Banerjee and Panicker, 2021), moral issues such as trust and deception (Dunn, 1991), technical and safety issues (Lee, 1993), and potential psychological and emotional risksincluding to themselves (McCosker et al. (2001).

Researching [with] children
Involving children and young people in research raises a number of practical challenges, particularly in relation to the gatekeeper's role. Gatekeepers are adults who are able to grant access or limit researchers' access to the participants (De Laine, 2000). They have a positive function in ensuring that children and young people are protected from research that could potentially be exploitative, invasive or coercive (Royal College of Paediatrics and Child Health Ethics Advisory Committee 2000). A study can have diverse gatekeepers drawn from different sectors, such as schools (Heath et al., 2007); government departments (Clark, 2010); health services (Heath et al., 2007), practitioners (Tidmarsh et al., 2003), as well as professionals, managers and workers who are embedded within institutions. Children and young people are often viewed as 'vulnerable', and in need of protection from harm, thus placing a responsibility on adults to protect their welfare. Gaining access to children and young people for research, therefore, often requires negotiation with multiple layers of gatekeepers at different stages of the research process, described by Coyne (2010) as 'a hierarchy of multiple gatekeepers' (p. 453).
Although gatekeepers help to safeguard the rights and interests of children and young people during research, Nutbrown suggests that research with children and young people has moved towards greater caution to reflect the 'riskier times'in which the research is being conducted, the result of which is a 'culture of restriction that pervades work with children' (Nutbrown 2010: p.5). Furthermore, Carter (2009) cautions that the practice of framing children within a discourse of vulnerability and seeing research with children as inherently risky invariably positions researchers as 'dangerous' the result of which is an over-cautious approach to research involving them (Campbell 2008). Stalker et al. (2004) noted that many of the difficulties they encountered in accessing hospitalised children and young people for their study related to the role of ethics committees in screening social research proposals. However, ethics committees are only one example of gatekeepers. Miller et al. (2013) reported that, negotiating access to children who were HIV+ was particularly challenging and permission was gained only after a series of negotiations in the form of letters, telephone calls and face-to-face meetings. Going into the field, the researchers were clear that special permission would have to be secured due to the children's HIV+ status. However, the researchers were wholly unprepared for the plethora of ebbs and flows encountered involving the 'corporate parent' and local partners concerned with the care and protection of these children, and how these negotiations threatened to derail the entire study, whilst necessarily protecting and safeguarding this vulnerable group (p.167).
The 'series of negotiations' described by Miller et al. (2013) could result in researchers abandoning their attempts to access children directly and seek their proxies (parents/carers) views instead (Coyne, 2010: p.454). Notwithstanding the fact that gatekeepers can, and do, whether wittingly or unwittingly, present challenges to the initiation and/or delivery of a research project, Cree et al. (2002) suggest their study would not have been possible without the active support of parents and professionals, although they question the extent to which this 'sponsorship' from the gatekeepers led to children participating who may not have done so. Gaining research access to children and young people is not unique to research on GBV as the need to protect and safeguard the rights and interests of children and young people transcends any subject or project.

Methodology
This article draws on a combined descriptive and autoethnographic research methodology. Descriptive research aims to provide a detailed and accurate picture of a particular situation (Neuman 2006)in this case, the field experiences of country teams involved in a Global Challenges Research Fund (GCRF) and University of Huddersfield sponsored project on GBV. An autoethnographic approach is also used with the aim of providing detailed accounts of participants' work from their own viewpoint. Autoethnography utilises data about self and its context, to gain an understanding of the connectivity between self and others within the same context (Ngunjiri et al., 2010). As a research method, autoethnography takes a systematic approach to data collection, analysis and interpretation about self and social phenomena involving self. Put differently, autoethnography is self-focused. That is, 'the researcher is at the center of the investigation as a "subject" (the researcher who performs the investigation) and an "object" (a/the participant who is investigated). Autoethnographic data provide the researcher a window through which the external world is understood' (ibid, p. 3). Together, descriptive research and autoethnography allow the participants to speak with an uninterrupted voice thus making their reflective accounts more authoritative (Etherington 2009). These approaches were also considered to be suitable due to the small number of cases involved. Using thematic analysis, the responses are presented below as unique, albeit overlapping case studies. Country teams responded to the same eight questions (See Appendix A for the list of questions posed).

Sample
Four countries were involved in the study, and all four country teams (India, Jamaica, Uganda and the UK) completed the questionnaires and provided reflections on the field experiences of researchers regarding access to children and young people. Although each country in the project had a team of between three and six researchers, each country, rather than participants, in this article is treated as a unique case.

Field experiences and reflections
Data was analysed using QSR NVivo. Thematic analysis was also adopted. Five themes emerged, including: accessing schools; approach to access; levels of permission-accessing participants; lessons learnt; and advice for other researchers.

Accessing schoolspermission issues
The experience of country teams in accessing schools across project countries ranged from easy to complicated. Although each country had a 'Ministry Engagement Plan' and a 'School Engagement Plan' as part of their project design, in reality, things did not always go to 'plan'. Whereas Uganda provided that 'access to schools was easy', for India and Jamaica, things were complicated by government bureaucracy. In India, for example the collaborating institution had to sign a Memorandum of Understanding with the State government as well as a permission letter from the Education Department of the Mumbai Municipal Corporation before permission was granted to approach schools, and in Jamaica two separate ministry approvals were required (Ministry of Health and Wellness and Ministry of Education, Youth and Information). Unlike the centralised education systems in other project countries, in the UK, school leaders are free to decide which research projects to support. This presented a unique challenge as whereas there was much empathy toward the project, many school leaders were hesitant to get involved, leading the country team to surmise that 'school access is arguably the most frustrating part of the project' (UK).

Accessing schoolsdesign issues
Permission issues aside, access to schools was complicated by the large sample size (India, Jamaica). Furthermore, given the fact that children and young people were to be drawn from a range of school types (public and private, religious and secular, rural/remote and peri/urban, etc.), this presented a further challenge on the ground. In the UK, for example although it was decided that only schools in one nation of the UK (England) would be sampled, after several failed attempts to onboard schools, the team realised it may be more feasible to focus on two main geographical regions.

Accessing schoolsinterests and agendas
Faced with a myriad of challenges, country teams soon had to 'change our strategies' (India). In each project country, teams realised it was necessary to establish and/or cultivate partnerships and alliances with agencies and organisations that showed an interest in, or that were sympathetic to their work. Although, these alliances could not bypass formal channels in India and Jamaica, they certainly helped to move matters closer to achieving a more favourable result, since alliance partners often had access to high-level decision-makers in government.
In the UK, such alliances were important to convincing and reassuring school leaders that country teams were not merely acting on behalf of a university, but rather that they had the confidence of 'trusted' organisations working in the area of GBV, organisations widely known to school leaders. There was another dimension to this which relates to a school's and/or leader's interest in GBV. For example 'the schools that were easy to access were the ones very interested in our work, and invested in the issue of GBV. Some teachers/administrators upon hearing the content of the survey were eager to help in its administration as well'.
'The schools that were difficult to onboard were the ones where they thought that this is not a topic to discuss with the children and were very wary of their academic schedule and burden of work' (India). This was also the case for nearly all schools in the UK. That is, schools participated where their interest in the topic outweighed their fear of any negative consequences such as 'a child disclosing something, and it ends up on Facebook' (UK).
In Jamaica, 'leadership of the school is important because if the senior management were receptive to us coming in and carrying out the study, from the point of contact, even to the point of facilitating the process when we got to the schools, that would impact everything'. In some schools, 'there were persons appointed to assist us, and that made the process smoother' (Jamaica). Juxtaposed against this, however, was the finding that 'if the contact persons that we had within the schools weren't interested, we got a bit of a run around, got delayed responses, or when we tried to contact them again, you know, they weren't able to provide us with any decisive response as to whether we could be facilitated' (Jamaica).

Approach to access
In the Ministry Engagement Plan and School Engagement Plan, country teams had three approaches to adopt: top down (via a relevant ministry or ministries); direct approach (approach schools directly); and alliance approach (in conjunction with another organisation). Each country had to adopt the approach or combine approaches in line with prevailing government requirements and culture.
For Uganda, 'access was easy because the team had the support of the authorities at both national and local levels'. At District (including Kampala City) and Municipality levels, specific letters were written to the head teachers instructing them to allow the team to engage the learners. According to the Ugandan country team, 'this directive made it easy for the research team to secure the cooperation and support of head teachers at selected schools'.
In India, 'the initial plan was to access an education conglomerate that had a fair number of schools across the city under its umbrella. Despite showing the Memorandum of Understanding (MoU) with the government of Maharashtra, the chairperson of one such conglomerate was not satisfied, which led to us making our own list of schools to reach out to'. Due to a delay from the chairperson, the team changed approaches and 'with the help of a market research agency that we used to recruit participants for the qualitative research… we reached out to some church-affiliated schools. The church-affiliated schools were somewhat easier to recruit because they were approached by one of our consultants who is a Christian and had previous associations with them, and by one of our research team whose relative was a trustee of a church organisation. Additionally, bearing in mind geographic diversity, we reached out to other schools by way of cold calling'.
For Jamaica and the UK, the primary access methods were top down (via a ministry) and direct (approach schools directly), respectively, and all project countries engaged in various degrees of 'cold calling' to schools.

Levels of permissionaccessing participants
Accessing permission from the authorities (ministries) did not automatically translate into automatic access to participants. There was a combination of bureaucracy, structural issues, communications, school management and related issues.
Team Uganda recounted a multi-layered process involving various actors depending on whether the children and young people that needed to be accessed were in city, towns or rural areas. 'At national level, we obtained clearance from the Uganda Ministry of Education and Sports (MOES) and Kampala Capital City Authority (KCAA) to access schools in the city. We then moved to the districts, where we approached the Chief Administrative Officers (CAOs) the heads of the technical wing at the district) for approval. The CAOs typically referred us to the District Education Office for guidance on how to proceed. The officials in the District Education Office were very helpful in the sampling process. For schools in the Municipal Councils, additional authorization was obtained from both the town clerks and principal education officers. From the local authorities the team directly approached the selected schools. Detailed explanation of the study was provided to the school administrators which they relied on to make a decision' (Uganda).
For India, 'the municipal schools required an additional permission letter from the Brihanmumbai Municipal Corporation's (BMC) "Public Private Partnership" department. The BMC schools were very strict about the permission letter. This delayed us further since reaching out to government offices and getting permissions were tedious tasks because of complicated Indian bureaucracy. The church-affiliated schools were somewhat easier to recruit'.
For Jamaica, 'once the approval was received from the Ministry of Education and the Ministry of Health, we would have gotten an approval letter. And that approval letter would have also been sent to the list of schools we had generated to participate in the data collection. And once that email was sent to the schools, our team would follow up with the school's administration. Typically, it involved the principal or the senior management within the school giving approval, and then the guidance counsellor would also be involved to oversee the process, the data collection process'.
In the UK, 'the process was less hierarchical and where school leaders accepted the invitation or request to participate in the project, responsibility was usually handed over to a named person who would serve as the main point of contact'.
In addition to apparent country and school protocols, structural (and infrastructural) issues such as lack of electricity, school telephones, computers and/or internet access all presented as additional challenges in some schools in India, Jamaica and Uganda. The 'general communication with some schools, being able to access accurate contact information for some schools, and also being able to actually reach the persons that we wanted, when we tried to make contact to those schools was sometimes challenging for us' (Jamaica). These challenges were compounded when the receptiveness and level of preparation of schools meant that they were a little bit more disorganised because the management team 'didn't put certain things in place to facilitate the process' (Jamaica).
In addition, 'timing was also a challenge. As you perhaps can imagine, schools have various activities happening at different points during the year and so depending on when we contacted them, and would have wanted to enter the schools, if they had activities happening during that time period, then it would have been a challenge to get access to those schools during that time. And of course, timing is not just the time of year, but also the time of day and the week in which we would have wanted access. If there were activities happening within those times, then it would have been difficult to access the students' (Jamaica). A school's teaching timetable and other school related activities and priorities also presented challenges for the UK and India country teams with the India country team reasoning 'the sample size of the survey was large and the conversion rate of the schools started dwindling as we progressed with each school'.
Absenteeism in India compounded the challenge of access to students since, 'the students who used to come on one day when we distributed the consent forms were not present on the day of the survey and vice versa' (India). Furthermore, 'time-management was another difficulty our team faced with the schools because sometimes students would take a lot longer to complete the survey than anticipated, which meant either they would have to miss the next class or leave the survey incomplete'. (India).

Lessons learnt: reflections on methods and process
As with any research project, there are always things that researchers could do differently. Country teams reflected on different issues from the approach to recruiting schools to participate, staff size and specialist skills, cost, time and access to rural schools, to the degree of power held by school leaders. For example India's country team commented, 'in terms of recruiting schools for the survey we took a trial and error approach; however, that required a certain amount of time and investment that we did not foresee. Especially because our team was very small and did not have necessary access to ministries we were learning and strategizing as per the situation and available resources. If we had a larger team with trained social work students from the beginning along with more time, we could have onboarded schools at workshops perhaps'.
Jamaica's country team reflected on design issues, compounded by costs. 'Looking back in hindsight, you know, it's always a different experience. So yeah, there are a couple things as a team, we discussed and agreed that we could have done differently, for example, it would have been really helpful if we had more field workers available from the start of the data collection period. But due to financial constraints, this wasn't possible. Had we had, you know, more field workers, then we could have gone out in teams, and done the data collection concurrently. This was something that was only possible later in the project'. Jamaica's country team also highlighted the importance of and need to have permissions and arrangements for the day of the visit reconfirmed before attending a school. They recounted, 'we would have turned up to schools that had already confirmed, only to find out that they weren't ready to receive us because maybe the forms weren't sent off, or some of them weren't sent. Or, you know, it wasn't a good time, because that particular day, some other activity was planned, or there were exams, or some of the students may have not been on site'.
The UK's country team reflected on challenges presented due to local management of schools. 'The biggest learning is undoubtedly around local management of schools and how school leaders play this out when it comes to research. Whilst we knew school leaders are the ones to decide on what projects their schools take part in, we did not anticipate the degree to which schools would opt not to participate. School leaders have enormous power, and although this is a very good thing, it did significantly hamper our field activities'.
The nature of questions in the survey, differences between public and private school engagement and the difference in school population (especially among rural schools) were key lessons learnt for the Uganda country team. 'The few schools that objected were under private ownership. They were particularly not comfortable with the questions on violence against children in school, probably because they would be exposed. Final exams were being held in some schools which made it hard for us… In addition, some secondary schools especially in rural areas… left the team with a small proportion of students'.

Advice for other researchers
The experiences gained by country teams through this project have provided them with several insights, many of which they feel will be of benefit to others researching sensitive topics or researching the same target population in different countries.
The India country team suggested 'first is to have enough backups of schools to reach our target numbers. Secondly, fieldwork in India is a complicated process and requires dedicated workers whose primary focus should be on strategising/liaising with not only institutions but also ministries and individuals who occupy positions of power. Thirdly, understanding that every sector has its own set of gatekeepers and in the field of school education it is important to recognise the work of NGOs and CBOs and form alliances with them as well. Fourthly, when working with groups who do not have experience in fieldwork for quantitative research studies, it is essential to have capacitybuilding training especially when the concerned issues are sensitive in nature. For this, we partnered up with an organisation that gave the larger team a clear understanding of laws/bylaws and systematic approach to working with young people. Fifth, access is only achievable when we are able to inform the gatekeepers about the relevance of the issue; however, sometimes it is impossible for researchers to get access because of structural issues and because the people in power are disinterested in accepting change in the status quo. Knowing where to commit and where to let go is pivotal in gaining access to institutions'.
The Uganda country team highlighted: 'it is important for the researchers to obtain clearance from the responsible authorities at national and local levels, as this eases entry into the schools. Secondly, building relationships with the school administrators in advance is critical. It is also crucial to identify a contact person for the study at the school level. However, most schools in Uganda are understaffed, the teachers are underpaid and demotivated. Therefore, most of them are usually unwilling to take on any extra duties. Planning for ways to compensate for their time can help to boost their morale to support research activities'.
Insights provided by the Jamaica country team included: 'first thing I would say is that if you have plans to conduct research on vulnerable populations, such as children, you should anticipate wait time with getting permission or approval from the oversight bodies responsible for children…. So if you are planning to do research on children anticipate that, you know, from the time of submission to the time of wanting to start your research, there might be some lag time or wait time. Another factor that I think is another consideration is, when you're conducting research on a sensitive topic, such as abuse or violence, there might be some concern from where you want to access your sample from. We recall for some schools, they were perhaps not as willing to participate in the research, because of the sensitive nature of the topic and not wanting to be associated with this particular topic. Even though the research was anonymous, and the schools wouldn't be named, etc, there might be some cultural sensitivity around certain topics where a person simply does not want to be associated with that sort of issue. This may lead them to decline to participate or give access to the population that we want to access. Another consideration is the general level of mistrust from the population itself about who is conducting the research, who are the persons that are trying to access information about them? So like, for instance, we found that we had questions about whether this was UK based research or research being done by a Jamaican institution. And so we had to provide explanations around that. Because persons have a general mistrust about research, and this can be worse if researchers come from an international body. Another issue concerns informed consent from parents. We had the opt out option, which became a little complicated. And in our culture, in our context, we are generally more familiar with opting in. So persons are more familiar with giving verbal consent, rather than written consent…. So that became a bit complicated for us'.

Discussion
Three important issues emerged from the field experiences of researchers involved in negotiating access to schools on a global GBV project led by the None in Three Research Centre at the University of Huddersfield, UK. The issues were gatekeeping, power and research management; safeguarding children and young people and, cultural competence.

Gatekeeping, power and research management
A relationship based on trust in a research process is vital to the success of any research study, and it is often gatekeepers who, through their actions, try to establish this trust. As set out earlier by De Laine (2000), gatekeepers are individuals, groups and organisations that act as intermediaries between researchers and participants. In this study, and beyond ethical review boards, there were not only gatekeepers, but also multiple levels of gatekeepers (national/government education and health ministries, regional bodies, school leaders, teachers, parents, etc.) described by Coyne (2010) as 'a hierarchy' (p. 453) with each exercising power and control over access or access points. Delays caused by gatekeepers in the four project countries created knock-on effects for the entire project schedule. In Uganda this was not as acute as in other project countries with the 10,000 students required for completing psychosocial surveys actually doing so. However, in India the numbers were 8365, in Jamaica the numbers were 7182 and the UK just under 3000 students participated.
In addition to controlling or managing access to a potential research site, gatekeepers may provide backing for their research project (Clark, 2010;Corra and Willer, 2002). This was strongest in Uganda where the education ministry directed regional authorities and schools to engage with the project. Similarly, in India and Jamaica, where the research teams had befriended the contact person, or where the contact person was sympathetic to the topic, this often resulted in the that person serving as an informal 'project champion' and in helping the researcher to navigate institutional factors and to secure engagement or buy-in. This power dynamic is important and may not always be a negative one, described by Cree et al. (2002) as a form of 'sponsorship', which was played out numerous times in each project country. The active support of 'gatekeeper-sponsors', without whose advocacy and championing, participation rates in the project countries in this study would have been very different was essential. Banerjee and Panicker (2021) also described the enthusiasm of children and young people to get involved in the study in India.
Despite advanced and careful planning by the country teams, gatekeeping issues exposed the vulnerability of research teams and their reliance upon individual interests and goodwill. Miller et al. (2013) noted they '… were wholly unprepared…' (p. 167) for the access issues that confronted them in the field, and a similar conclusion has been reached by the UK country team which described 'the biggest learning' regarding access issues being '… the degree to which schools would opt not to participate. School leaders have enormous power, and although this is a very good thing, it did significantly hamper our field activities'. Put differently, very little advanced planning can adequately prepare you for some field experience, including the 'suspicion' felt by the Jamaica country team from some prospective participants who demonstrated mistrust about international researchers. Furthermore, gatekeepers may have found it difficult to engage with the research team due to lack of time, resources and disruption to their work and/or their organisation (Clark, 2010), since locating information, providing links, answering queries and approaching participants diverts resources away from the researcher's central role within his or her organisation and increases workload (Din and Cullingford, 2004) although this was not something raised by country teams.
It is to be understood that gatekeepers have a duty of care to research participants and any potentially problematic outcomes for them need to be mediated and resolved before access is granted (Miller et al., 2013(Miller et al., ). 1996. For example when the gatekeepers demanded additional permissions (Jamaica and India), and when the India country team had to sign a Memorandum of Understanding with the government/education ministry before being granted access to the children and young people, possibly for legal concerns (Munro et al., 2005), potential issues of ethics, confidentiality and representation (Banerjee and Panicker, 2021), unwanted intrusions (Curran & Cook, 1993), privacy concerns (Murray, 2005) were being mediated.

Safeguarding children and young people
All four project countries are signatory to several international conventions and treaties directed towards safeguarding and protecting children and young people. Accordingly, 'what to the researchers felt like them being given the run-around was in fact calculated steps being taken by government and partner agencies to secure the human rights of this vulnerable group' (Miller et al., 2013: p.172). Governments and other gatekeepers do not want to risk the welfare of children and young people to researchers no matter how important a research topic may be or may seem to the researcher. In each of the four project countries there were concerns about disclosure, confidentiality and 'something' gaining unwanted or unwarranted media attention (including social media). Gender-based violence can bring feelings of shame for victims and the welfare and stigmatisation faced by victims was a matter of serious cause for concern for both country teams and gatekeepers alike. It is perhaps understandable then that in trying to safeguard the rights of children and young people as established in Article 17 of the International Covenant on Civil and Political Rights (ICCPR), Article 16 of the Convention on the Rights of the Child (CRC), Article 12 of the Universal Declaration of Human Rights (UDHR), and relevant national legislative instruments, gatekeepers acted in the best interests of children and young people by attempting to minimise potential risk to them through scrutinising and interrogating the research team before access, and even afterwards, as in the UK where access was withdrawn after the questions posed in the survey caused 'upset' to a young person.

Cultural competence
A project of this nature and size also has important lessons for how researchers engage with their own or with different national and institutional cultures. Although the ministry and school engagement plans were conceptualised and designed by lead academics from the University of Huddersfield, UK, country teams had discretion to adjust and amend these to suit local contexts. Whereas the principle of these plans was consistent with standard research ethics for field engagement (Morrow and Richards, 1996), there may well have been an over-reliance on the country team's insider status and presumed knowledge of context. This is important for two reasons. Firstly, knowledge of context does not necessarily translate to knowledge and understanding of processes and cultural practices around research, especially researching certain groups, including children and young people. Secondly, a country's and/or an institution's processes are part of its culture, and country teams may not have paid enough attention to both the formal and informal practices and behaviours unique to their local country context and to the institutions within each country. This is an area that is worth further investigation. Suffice to say, however, for example in Uganda, although the education ministry issued a directive, there was a tall hierarchy of agencies and persons who had to be satisfied before access was granted. In India, the country team did not expect to sign a MoU with the education ministry/government. In Jamaica, the country team did not expect to have to gain permission from both the education and health ministries. And in the UK, the country team assumed local management of schools would have made it easier to onboard schools due to localised decision making powers. This suggests that country teams did not fully appreciate how national and institutional cultural practices and processes can and do play out in researching sensitive topics with children and young people.

Conclusion
Involving children and young people in research is as important as it is problematic and involving children and young people in research on sensitive issues is even more problematic. In national, institutional and cultural contexts where discrimination and stigma relating to certain topics may be significant, this can be even more challenging. Gatekeeper power, safeguarding children and young people, and understanding how processes and behaviours at national and institutional levels influence the research process, including research led by or involving international researchers are key reflections in this article. Scrupulous research planning and management are no substitute or match for having an understanding of how systems and processes work in a country, and in institutions within that country. Thus, having a knowledge of a country and the institutions therein, or being located in a country should not be equated to having knowledge of the processes, practices and behaviours (culture) of the country itselfas far as access and gatekeeping issues are concerned.
Gaining and retaining access to participants in project countries was a continuous exercise in charm, negotiation, perseverance and determination, underpinned by 'a belief rooted social transformation' (India). The challenges of the field invariably tested the relationships among country teamscausing severe stress and anxiety, (and these effects too, are to be acknowledged). Nevertheless, 'the difference of perspectives among the research team and the institutions/ministries eventually also helped us learn that we cannot lose hope and enabled us to keep working on targets that were realistic and achievable' (India). Furthermore, 'taking time out to talk, to bear each other up, and to regroup' (Jamaica) was another way to re-centre relationships and the aims and ambitions of the project. Country teams agree that gatekeepers did not act out of spite or malice towards themselves and/or the project, but instead were acting in the spirit of the Convention on the Rights of the Child (1989), and therefore in the best interests of children and young people.