Data Bites – Introduction to Programmatic Data De-identification with R

Workshop: Programmatic Data De-identification with R

This practical workshop, delivered by the UBC Library Research Data Management team, introduces programmatic approaches to de-identifying sensitive research data in R. Through hands-on exercises using a realistic survey dataset, participants will apply a structured workflow, from assessing privacy risks to exporting a shareable, de-identified dataset.

Participants will learn how to:

• Identify privacy risks in research data, including direct identifiers, dates, geographic variables, and free-text fields.
• Apply de-identification methods in R using dplyr, including removal, generalization, suppression, anonymization, and pseudonymization.
• Run quality assurance checks to confirm a dataset is sufficiently de-identified before sharing.
• Export a de-identified dataset and a data key file, and understand best practices for securely storing each.

—

Note 1: Participants familiar with R and RStudio are welcome to follow along on their own workstations. If you are new to R, we recommend watching first to learn how R can be used for analysis and data de-identification, then practicing later with the resources introduced during the session.

Date: Wednesday, July 15, 2026

Time: 10:00am – 11:00am (British Columbia)

Presenters: Eugene Barsky, Vanessa Choy, Grigory Artazyan