Paper short abstract:

Ethnologists and folklorists use a range of strategies to share the data they collect while protecting individual identities and respecting community preferences. Could a novel approach to analysis developed by biomedical researchers help us to responsibly open up our own fields?

Paper long abstract:

Ethnologists, folklorists, and scholars in adjacent fields have long sought to withhold portions of the data they collect from public circulation, in order to avoid disclosing the identities of their interlocutors and to respect community preferences. This field-level norm poses a challenge for open science, which offers the prospect of more robust and collaborative inquiry facilitated by the sharing of research data. Strategies for sharing qualitative data in keeping with these restrictions have included reprocessing data to conceal certain details or appointing a gatekeeper to filter data requests by their purpose and the social location of their initiator. How should ethnologists and folklorists think about the relative merits of these strategies? And could digital innovations in other fields help us to share our data in a responsible way while also advancing possibilities for meaningful reuse? This paper outlines an approach known as nondisclosive federated analysis, sometimes characterized in terms of "taking the analysis to the data, not the data to the analysis." Here, a researcher initiates a query that is routed to different servers where datasets of interest are securely stored, and then pooled results are returned to the researcher under certain parameters that prevent identity disclosure. The approach was developed in the context of computationally intensive biomedical research, but this paper explores whether and how it could be adapted for use with qualitative data, addressing issues like the need for ongoing markup. It concludes with a call for collaborators who might be interested in testing the approach's viability.