Examining How Open Research Affects Vulnerable Participants
Three anthropologists, a psychologist and a social inequalities researcher walk into a seminar room… It’s not the set-up for a joke. It is the set-up for a thoughtful discussion about the role of open research in qualitative research with marginalized and vulnerable communities. This is precisely what was delivered in a recent LSE Open Research Working Group event: “Making research on vulnerable or marginalised communities open.”
Conversations around open research are often driven by the hard sciences or quantitative social sciences, yet this obscures important open practices by qualitatively focused fields like anthropology. This event revealed important insights into what openness really means, beyond the straightforward sharing of data, methods and publications. It revealed how openness can be embedded at a small scale through participatory translation practices, how it is reflected in the transparency of revealing the researcher’s positionality to their participants, and how research can be more open to broader society by collaborating with organizations.
Varying kinds and degrees of openness
Open research methods are sweeping Higher Education institutions with promises of greater transparency, rigour and trustworthiness in research. However, as many of these methods (such as preregistration, data sharing and reproducible code) are derived from quantitative scientific methodologies and concerns, this creates dilemmas for qualitative researchers, especially those working with marginalised communities. Data sharing may not be possible where it endangers oppressed people, or where even storing such data places the researcher in danger. There are also questions about how relevant concepts like replicability are for research methods that focus on particular contexts, or whether preregistration is useful for ethnographic researchers whose “sample” and research questions change in the course of fieldwork.
The degree of how open qualitative research can be, as well as what kind of openness is appropriate, rests on the research context. For example, Jo Hemlatha did not anonymize most sex workers involved in their research. This was because the sex workers had been politically asserting their identities for decades before and not identifying them would do a disservice to their work. For Polly Vizard, however, a lack of guidance stopped her team making interview transcripts from CASE publicly accessible, even though reports have been open access for 25 years.
George Kunnath cannot consider publishing interview transcripts from his work since a lot of the research with guerilla groups cannot be written down or recorded due to ethical risks that participants face from state actors. However, more recently, George has sought linguistic and attributional openness by co-authoring in local languages with local academics who are comfortable to do so.
These are just some examples of how open research can be varied in approach depending on participant safety and friendship between researchers and interlocutors. This brings us back to the question of how open something must be, and the meaning behind openness as understood by different people and communities. Openness is also about responsibility – who do researchers feel responsible towards? The communities we work with, or institutional understandings of ethics and openness which may not encompass the breadth of issues multiple communities may face with “one size fits all” policies? We suggest that the researcher’s primary responsibility is, in most cases, to the participants they work with above the institutional policies.
Practices of openness
Openness as a practice is a consideration many researchers have made, especially those working with marginalized communities in anthropology, gender studies, public health, participatory research and action research (to name a few). Qualitative researchers at large incorporate open research practices into their work, through participatory methodology and ethnographic methods. They have done this through a range of practices not commonly found in guidance on open research, but which nonetheless promote the Open Research values of transparency, integrity, honesty, and accountability. For example, filmmaker Kat Mansoor and anthropologist Andrea Cornwall made the film Save us from Saviours with the advice of the sex workers it features, with filmmaking being the sex workers’ preferred way of telling their story.
Two important methods which qualitative researchers of marginalized communities can offer to open research are participatory methods, and alternative methods of public dissemination. Working with “experts by experience” is an important aspect of openness for any social scientist, as it facilitates the relevance of research for the people about and for whom it is made. Public engagement also promotes openness by making research more accessible and understandable to the public. From within the panel, Myofora Kakoulidou works with “experts by experience” across the research life cycle – from formulating research questions to dissemination of findings. Similarly, Jo Hemlatha uses extended meetings and discussions with participants, and Polly Vizard described her “good practice model of continuous involvement with a civil society organization.” All of these methods focus on the involvement of traditionally marginalized voices, and opening the agency these voices have over research with them.
Perhaps the most overlooked open research practice is interpersonal openness. This is an attitude of transparency about the aims and process of the research project, and about the researchers’ motivations and backgrounds – their positionality – when communicating with research participants. For example, George Kunnath discussed how co-authorship with participants creates transparency about who creates knowledge and makes the research transparent about its partiality, rather than feigning neutrality. Jo Hemlatha similarly talked about how openness and participation begin well before the research project itself—intentions and motivations must be reflected upon so that the project design stems from a genuine desire for openness, rather than treating it as merely a tool to measure impact.
Barriers to qualitative open research
Open research, according to the dominant positivist paradigm, presents several barriers to researchers using qualitative methods, especially where such methods concern marginalised communities. Such barriers demonstrate why alternative practices and ways of viewing Open Research are necessary.
When researching vulnerable groups, there exist risks of stigmatization, identification, and extractivism that lead to welfare impacts and exposure to hate crimes or violence. Open research must therefore be understood through continuous and long-term communication and engagement with the community. Openness for one may not mean openness for another. Anthropological and participatory research practices could help indicate these differences early on, so that researchers can have conversations on what openness offers different groups.
Research is also messy – there are often conflicts between subjective experiences within a community and the range of acceptable ideas that are reproduced collectively. This makes open research a dynamic process and challenging to ensure transparency and rigor. George Kunnath says on the messiness of using anonymization practices, that “you can anonymize people but you can still identify them”, but as Jo Hemlatha says, often the only way to understand, investigate and use this inherent messiness is to “get dirty.”
A lot of open science is still in English, and behind technological barriers. Sometimes, even though outputs are made open after the project ends, the project itself may not have used open, participatory methods of engaging with interlocutors. Using participatory methods from the beginning of the process invites the engagement of interlocutors and opens the research process to them early in the research. Ensuring the language in which one publishes, and methods of research dissemination are also accessible to a wider audience (beyond dense language, Anglo-centrism and scriptocentrism) further opens research up to communities and researchers worldwide. Finally, using technology to widen access – beyond paywalls, through multimedia and spaces for open discussion, Open Science can be understood and used more broadly but also locally, making it “as open as possible, as closed as necessary”.
