How BIRI Works

BIRI enables researchers who wish to study the effectiveness of their intervention approach to support students in higher education at scale. The intervention format is flexible: an intervention can consist of text (e.g., study tips, stories, testimonials), images (e.g., infographics, photos), videos (e.g., a tutorial), and writing activities (e.g., to write about purpose, utility value, or belonging), as long as the intervention content can be placed in one or two stand-alone modules that are independent of the rest of the course content.

Our team aims to create a cross-disciplinary research environment that brings together scientists in education, psychology, computer and information science, behavioral economics, and the behavioral and computational social sciences. BIRI will enable robust, open, and iterative research by facilitating intervention implementation, data collection, and study replication. This will dramatically reduce the time it takes for intervention researchers to conduct research while increasing the number of researchers who have access to experimentation sites in education.

BIRI uses the following steps for each proposed intervention project:

  1. Researchers submit a study proposal
  2. Course teams submit course information for courses that are offered using the Realizeit adaptive learning platform
  3. We match researchers with relevant courses based on the information in the proposal
  4. We add the proposed intervention content as a randomized controlled trial into one or more courses
    Optional: Researchers can add additional questions to the standardized course survey
  5. Run intervention once (and replicate)
  6. Provide de-identified data to researchers: affect, survey, behavioral, performance data

Researchers are encouraged to pre-register their studies before they receive the de-identified data.

Affect detectors are being developed and validated for Realizeit courses within this project, and will provide fine-grained data on this aspect of the student experience.