- LearnSphere, a new federally-funded tool from Carnegie Mellon University, aims to build a large repository of data on student performance, with the goal of analyzing how students learn.
- The project shares similar goals with InBloom, the database of longitudinal student data that shut down last year after districts backed away over student privacy concerns.
- LearnSphere’s founders say they’ve dodged several of InBloom’s student privacy pitfalls.
Ken Koedinger, LearnSphere’s project leader, told The Hechinger Report that “there are certainly some similarities” between the companies while cautioning that there's also “a deep philosophical difference.” LearnSphere will look primarily at students’ key strokes and won’t touch school records or personal information. What will they leave out? Student names, addresses, zip codes, social security numbers, race, family income, special education designation, and other student identifiers. Koedinger is very firm on that final point: “The student identifier column, even if yours is already anonymized, we re-anonymize it automatically,” he told The Hechinger Report
The data will also be less centralized. Schools and districts will control access and have the opportunity to upload the data they choose. Once they’ve done that, a research manager will monitor data to make sure nothing that could identify a student is included.
So what could it be used for? Koedinger sees its value in determining which teaching strategies are the most engaging and effective, from videos to reading assignments.