Big Data and Data Science Research Ethics

This module introduces learners to big data, data science, and research ethics. It discusses how learners can think about data science ethics as a practice best learned over time.

Recommended Use: Required
ID (Language): 20986 (English)
Author(s): Katie Shilton, PhD - University of Maryland; Emily Dacquisto, MSW - University of Maryland

Big data creation, collection, and analysis pose multiple challenges to privacy. These include inference risks and reidentification risks. To mitigate privacy challenges, researchers have multiple options, all of which have benefits and tradeoffs. This module introduces several ways of thinking about the privacy challenges of big data and leads learners through various privacy protection techniques and tools for big data research.

Recommended Use: Required
ID (Language): 20987 (English)
Author(s): Katie Shilton, PhD - University of Maryland; Emily Dacquisto, MSW - University of Maryland

In the U.S., research ethics has long relied upon several components, of which informed consent may be the best known. Informed consent means disclosing understandable information that participants can use to decide whether to participate in research and making it clear that participation is voluntary. This module explores the tension between participant expectations for research participation and the reality of research needs and regulatory boundaries. Within this gray space, researchers must make decisions about when and how to ask for consent (even if it is not strictly required), ways to engage with research communities beyond traditional forms of informed consent, and practices like broad consent that can improve participant trust.

Recommended Use: Required
ID (Language): 20988 (English)
Author(s): Katie Shilton, PhD - University of Maryland; Emily Dacquisto, MSW - University of Maryland

Privacy protection techniques and methods of facilitating participant awareness are important pieces of research ethics for big data and data science, but they do not completely address participant concerns with data science methods or the ethical obligations for using big data. Data scientists must also grapple with the fact that data can be a tool of power (the ability to influence people’s ideas, actions, or behaviors). As this module explores, the uses of data and quantification to make decisions about people have not always resulted in empowerment and flourishing but instead in marginalization and oppression.

Recommended Use: Required
ID (Language): 20989 (English)
Author(s): Katie Shilton, PhD - University of Maryland; Emily Dacquisto, MSW - University of Maryland