Guidelines for data management and scientific integrity in ethnography

Abstract

New protocols for scientific integrity and data management issued by universities, journals, and transnational social science funding agencies are often modelled on medical or psychological research, and do not take account of the specific characteristics of the processes of ethnographic research. These guidelines provide ethnographers with some of the most basic principles of doing such research. They show that the primary response of ethnographers to requests to share research materials with third parties should be to remain aware of the fact that these research materials have been co-produced with their research participants; that the collaborative ethnographic research process resists turning these materials into commodified, impersonal ‘data’ that can be owned and shared publicly; and that therefore the primary response of ethnographers should be to retain custody of research materials.

Social scientists are increasingly confronted by attempts of employers, media, policy-makers, funding agencies and journals to regulate their management of research materials through protocols developed for sciences that employ a formal, context-neutral design usually borrowed from medical research. These protocols do not recognize the specific nature of qualitative social science research, or regards it as exceptional or problematic. As a result ‘data management’ may effectively hinder responsible research conduct and threaten the scientific integrity of ethnographers. The following guidelines sketch the basic scientific features of ethnographic research, and outline why the handling of research materials in qualitative social science requires paying more attention to the process and the establishment of responsible social relationships during research. They are adapted from a statement about data management and scientific integrity developed by Leiden anthropologists, that was recently published as a Forum Discussion in Social Anthropology/Anthropologie sociale. ¹

Ethnographers recognize that social research is necessarily rooted in social relationships. The social relationships built by ethnographers provide a qualitative, intersubjective and value-laden foundation for knowledge. This knowledge usually derives from the mutual co-productionof research materials involving both researchers and researched. ² This implies that research materials or data are rarely fully owned by either researcher, researched, let alone a third party. The first duty in ethnographic research is therefore to recognize this joint production and joint ownership of research materials. All forms and norms of managing data depend on it.

The collaborative nature of ethnographic research implies, secondly, that researchers should continue to treat research materials and data as collaborative for as long as they work with them. Ethnographic research is a process, in which the establishment of trust and the interpretation of data continue to evolve, as mutual understanding of both researchers and research participants changes in intensity and meaning. ‘Data’ are therefore never completely fixed and finished products, nor is consent ever completely informed by the quasi-contractual gesture of a written consent form. Prior and written consent provides only an artificial ethical security in the ethnographic research process, and may be deceptive towards research participants: consent forms can never predict all contingencies of the research process and may even themselves threaten to disclose data that should be kept private. Moreover, ethnographers have the duty to keep in mind that the commodification of research materials as ‘data’ may obscure questions of intellectual and cultural property. Ethnographers have long preferred the dynamic possibilities for renegotiation that oral forms of consent allow.

Protocols for data management, moreover, rarely acknowledge the decisive importance of the processingof research materials into data. While data in hypothesis-testing and experimental research commonly implies some form of anonymization, quantification and the disguise of research participants’ identities, the interpretation of case studies in ethnographic studies often resists such commodification and makes use of personal, intimate and sensitive information. ³ Processing research materials to the point of making participants unidentifiable makes much written and voice-recorded materials virtually useless to third parties, and is impossible in all research that involves photography or film. Moreover, the use of protocols for data management may hamper the long-standing ethnographic research engagement with communities and individuals that are vulnerable, rendered illegal and/or subject to discrimination, because formal registration of access to data and prior consent may contradict the conditions for guaranteeing confidentiality and trust to participants.

The third necessary condition for ethnographic research is that individual researcher(s) can and should be responsible for the integrity, preservation and protection of the materials gathered during a specific research project like any other caretaker of collective property or disciplinary standards. ‘Researchers have an ethical responsibility to take precautions that raw data and collected materials will not be used for unauthorized ends’ and this includes ‘establishing by whom and how records will be stored, preserved, or disposed of in the long term’. ⁴ In most cases, this does not depart from the common practice by ethnographic researchers to keep records in their personal custody and possession, to protect them by passwords, and to decide on a case-to-case basis whether data can be shared with third parties. In the current climate of responding to protocols for digital sharing of data with other scientists in repositories, this involves adopting some form of lasting embargo on a substantial part of the research materials for reasons of privacy and intellectual property. Therefore, even when employers claim ownership of these materials, this implies that their access to or use of ethnographic materials needs to be conditioned by restrictions observed by the primary researcher(s).

In cases where scientific integrity is in doubt, or PhDs have to share data with their supervisors or PIs, access to data can be granted by the primary researcher(s) on a fully confidential basis. In cases where access to data risks inflicting harm on research participants, employers and professional associations should actively support some form of informant confidentiality comparable to prerogatives valid in journalism. ⁵ The collaborative nature of ethnographic research materials implies that ethnographers have a special duty to consider requests by research participants (or their descendants) to share data, unless this actively and unnecessarily harms (some of) them. In the current climate of ‘big data’ and the vulnerability of digitally stored data to hacking by outsiders, ethnographers should pay particular attention to both the ethical standards that pertain within social scientific disciplines and the considerations outlined by the Association of Internet Researchers. ⁶ In the case of the storage and sharing of audio-visual materials and material culture problems of identification and privacy or (object) ownership and repatriation may occur. For this reason, ethnographers should be particularly alert to the demand of constant renegotiation of informed consent, both when publicizing audio-visual materials and while managing the possession, storage and display of objects. Finally, ethnographers should be aware of and guard against the tendency to use data management protocols primarily to safeguard the (public relations) interests of employers and home institutions, rather than to contribute to social scientific knowledge and prevent research participants from (individually) coming to harm. Where this will not violate the above principles of ethnographic integrity, ethnographers also have a duty to consider making their research materials publicly accessible in appropriate ways.