Kristi Thompson

Research Data Management Librarian

Kristi Thompson

Kristi Thompson is the Research Data Management Librarian at Western University, a member of Western’s Non-Medical Research Ethics board, and chairs the Sensitive Data Expert Group of the Digital Research Alliance of Canada. She has many years of experience working with data as a librarian, researcher and statistical analyst. She’s published articles and book chapters on sensitive data and other elements of research data management, worked with funded research teams to support deidentification and publication of research data, and has spoken to national and international audiences on data anonymization. Kristi also co-edited the recently released OER textbook Research Data Management in the Canadian Context: A Guide for Practitioners and Learners.

All Sessions by Kristi Thompson

  • Tuesday, April 30


  • 13:15  - 14:00
    SESSION KEYWORDS: Research Security Data Ownership, Management and Governance

    SESSION ABSTRACT

    In 2021, Canada’s three federal funding agencies released a policy on research data management, including a mandate that grant recipients are “required to deposit into a digital repository all digital research data, metadata and code” that support their research outputs. This is not a requirement for open sharing, but most academic and journal data repositories are designed with open sharing as a default, and the wording strongly encourages sharing where ethical and other requirements allow. But when can data on human participants be shared safely? Many researchers promise research participants that they will only share data that has been anonymized, or even say that the data they collect is anonymous because no identifiers are collected. But when is this true? How can a researcher, a research ethics officer, or a data curator tell when data may pose a risk to confidentiality? Survey data often contains hidden indirect identifiers that could be exploited by a skilled attacker to determine a participants’ identity – anything from a date to a distance to a selected interview language could, in the wrong context or combination, pose risk. In this workshop I will introduce some of the theory that explains why data anonymization is so difficult, discuss techniques for assessing the risk in a dataset, and share some of the data anonymization issues I have encountered as both a data curator and a REB member.
    Sessions




Share

Kristi Thompson