Draft of Good Practice guidelines in collaborative research
‘data’ ownership, authorship and power, by Alice Tilche and Rita Astuti
The following text, written by Alice Tilche and Rita Astuti, is a proposal for guidelines for those involved in research collaborations, which emerged from a series of workshops and discussions held by Tilche and Astuti over a period of time. We are publishing their text here to begin a conversation about this topic and to encourage further discussion and reflection amongst EASA membership. Feedback and suggestions are most welcome (by emailing ethics(at)easaonline.org) and will be fed into a final version of these guidelines for approval at the 2020 EASA AGM.
These guidelines respond to the emergence of new forms of knowledge production, which privilege large, externally funded projects. In anthropology, the shift towards large projects is making new collaborations possible, opening up new modalities of knowledge exchange. However, it is also creating new forms of exploitation, precarity and hierarchy, potentially calling into question the key principles of our discipline. Collaboration is often organised through vertical rather than horizontal structures, with an unequal distribution of rights and responsibilities – for example, between those who gather the data and those who analyse it – that has negative repercussions for those structurally most vulnerable, while adding pressures on those at the top. Furthermore, the transfer of an ill-fitting science model to anthropology and the commercial pressures on anthropology departments, are turning scholars into producers of outputs, a redefinition of their role which can be seized upon by line managers and auditors, but also by senior scholars. This is detrimental to the intellectual freedom and career development of young researchers, and threatens the ethical integrity and epistemological underpinnings of a discipline that claims a particular relation to its ‘data’ (see below).
Recent guidelines published by the journal Ethnography and endorsed by the European Association of Social Anthropology (EASA) and the Dutch Anthropological Association, highlight broad principles for anthropological research, data management and scientific integrity in anthropology. The key principle they establish is that anthropological knowledge is always co-produced, embedded in particular social contexts and, as such, cannot be transferred to third parties without consent or consultation (Koning et al. 2019; EASA 2018). These documents are mainly directed at the attempts by others (employers, the media and policy makers) to regulate anthropologists’ work in ways that are epistemologically counterproductive and ethically problematic. Granted this, we must now extend such a line of reasoning to collaboration amongst anthropologists. With the nature of academic employment becoming increasingly precarious – for postdocs and temporary teaching-only staff, who must fight hard for permanent positions, but also for permanent members of staff, whose career progression is linked to the external auditing of an increasing number of metrics – it is not only ‘others’ but anthropologists themselves who are adopting questionable research strategies. These include treating fellow anthropologists as data collectors, using their research material without prior consultation or denying them access to it, as if anthropological data can be detached from the social relationships between researchers and research participants that have co-produced it.
While project-based research has become normalised, there are as yet no collegial, discipline-specific agreements on how this type of collaboration should be managed. These guidelines seek to clarify the relationship between data production, data ownership and authorship in anthropology and to suggest some ground rules for ensuring fair working relations within projects. We hope that they will inform revised ethics guidelines for the discipline and be used as a basis for negotiations within collaborative project teams and between project teams and funding bodies. The document is divided into two sections: the first is specific to anthropology; the second addresses broader issues of career development and project structure applicable across the social and natural sciences.
Key principles of research integrity
As already established by the EASA statement on data governance for ethnographic projects (EASA 2018), by Pels (2018), by Dilger et al. (2019) and Koning et al. (2019), anthropological research materials cannot be treated as disembodied and transferable data. They are always co-produced through relations of trust (between the researcher and her interlocutors) and through the interpretative work of the researcher (Koning et al. 2019: 2).
Consequently, in the context of collaborative research projects, every project participant ought to be considered as the guardian and guarantor of the integrity of the research material they produce and of its interpretation.
This has a number of implications:
- Members of a research team – including Principal Investigators (PIs) – cannot claim ownership of the research materials co-produced by others.
- Similarly, no member of a research team can use research materials without prior consent of, and consultation with, the research team members who produced them. This applies to qualitative, quantitative, experimental and visual data as well as interviews. Fieldnotes in particular should always remain under the guardianship of their authors and the sharing of field notes within a project should not be assumed.
- Every member of a research team must retain access to, and the right to publish as a single author from, the research materials they have co-produced (subject to collaborative arrangements they have established with their interlocutors), and they should continue to do so after the end of any given collaborative project.
Given the principles outlined above, researchers in anthropology are never to be treated as ‘data collectors’. Instead, they should have an automatic claim to the authorship of any publication that utilises their research materials.
This principle, when applied to the publishing model currently prevalent in anthropology, has a number of implications:
Every member of a research team has the right to appear as the author of publications that draw on the research materials they have co-produced. This right to authorship is based not just on their contribution to the writing itself, but to the process of co-production and interpretation of the research materials. Such contribution gives the right of single authorship, whenever a member of the research team is entirely responsible for the writing of the piece, and to first-authorship when they are involved in co-writing it (when multiple project participants contribute their research materials and are involved in the writing, they should all appear as equal co-authors).
When publishing as single authors, members of a research team must acknowledge the contributions of other team members to the research design and implementation, besides acknowledging the research grant that made the research possible.
Giving assistance during research design and implementation or providing feedback on draft publications do not grant members of the research team, including the PIs, automatic right to appear as co-authors. For members of the research team to use and write up unpublished materials and analyses that emerge from fieldwork other than their own, they must work closely (in the conceptualisation and writing of the piece) with the member of the research team whose materials and analysis they are drawing upon. As stated above, the latter has the right to authorship.
Co-authored publications should only be based on genuine collaboration, i.e. collaboration that is mutual, ideally all the way from design to fieldwork and from analysis to writing. This means that all authors should make a substantial contribution, i.e. bring comparative research material, and/or core methodological and conceptual insights to the publication. The order of authors in publications must fairly represent the contribution to the production of the research materials (i.e. actual involvement in fieldwork) and their interpretation.
Research materials should be published whenever possible in ways that fall under the aims of the project, but no members of the research team (including the PIs) have the right to control the interpretative work of others. This is crucial to guarantee the intellectual freedom of all researchers.
As noted, these guidelines are intended to apply to the current dominant publishing culture in anthropology, which prizes single-authored publications. This publishing culture is out of sync with the current funding model, which prizes large team grants.
For genuine collaboration to be possible, such publishing culture needs revisiting; alternatively, the preference for large collaborative grants should be abandoned.
While the preceding section has covered issues specific to collaborative projects involving anthropologists, the next one covers issues also applicable to researchers in other disciplines.
PIs should ensure that all members of the research team are given enough time and resources to pursue their career development. This is particularly crucial to early-career researchers who are typically in precarious employment.
We therefore welcome the recent recommendation by the Vitae Concordat Strategy Group that researchers should be allocated 20% of their time for their personal and professional development, e.g. for attending conferences, training or publishing independently of the focus of the project.
PIs and other senior members of the research team should prioritise giving support to early career researchers in producing publications, attending conferences, and engaging in other career-development activities, such as grant writing, that are appropriate to their career needs (this could mean single- or co-authored publications, depending on the discipline).
Institutional support for projects
Large, externally funded projects create new challenges for individual researchers and demand new institutional responses. In what follows we outline a few areas where intervention is needed.
Protocols for collaboration (e.g. co-authorship, data sharing, time management) should be put in place at the very beginning of any project, with the mediation of an external facilitator. Such a person should be familiar with and take into account the ethical principles and research protocols of each discipline. The presence of an external facilitator is crucial to ensure that the needs of all participants are taken into account. Should the ongoing nature of the project require a revision of such protocols, the external facilitator should be consulted again.
In order to prevent researchers’ isolation, projects should remain closely embedded in the life of a relevant academic unit (e.g. a department). Host institutions should ensure that academic units which host externally funded projects include researchers in that unit’s activities and offer them opportunities for intellectual exchange and development beyond the specific focus of the project.
Host institutions should provide PIs with initial and ongoing training on how to manage large projects (in the same way in which Heads of Departments are typically given training). This should include training on how to effectively mentor early career researchers and on how to create and maintain a healthy working environment in the face of the challenge of precarity. Host institutions should also facilitate dialogue among PIs from different projects (e.g. by holding regular meetings) to counteract PIs’ possible feelings of isolation and intimidation in the face of top-down demands from grant agencies.
Host institutions should provide training to all members of a research team on the technicalities of the grant (e.g. reporting mechanisms) so that everyone is clear about their rights and obligations.
All parties involved in a project should have oversight of the reporting process to project funders (e.g. by signing off interim and final reports) and should have independent channels with funders to raise concerns about the project.
Although PIs should provide early career researchers with ongoing mentorship, an external source of advice and recourse should also be available to them. Researchers should thus have a designated person/mentor within a relevant academic unit but independent of the project, who will act as an impartial source of career development review and advice.
Funding agencies should require host institutions to provide the support, training and mentorship just outlined. All these initiatives should be fully costed within grant applications. They should not become an added burden to either academic units or individual members of staff.
This document draws on a series of consultations at national and European level. At EASA 2018, Alice Tilche and Giacomo Loperfido organised a panel discussion (https://www.easaonline.org/conferences/easa2018/events) on the ethics and politics of big projects with the support of the PrecAnthro collective (https://www.facebook.com/precanthro/). In May 2019, with the support of the LSE Anthropology Department and its RIIF Fund, Alice Tilche and Rita Astuti organised a follow up workshop at the London School of Economics aimed at drafting a set of guidelines. We would like to thank the PrecAnthro collective which made this discussion possible in the first place, the Anthropology Department at the LSE and all those who participated in the workshops for their contributions. The guidelines are informed by the discussion that took place at the workshop, but do not reflect all the views expressed by the participants.