Conference Presentation, Lecture, Fireside Chat
Shriram Krishnamurthi: Curriculum Design as an Engineering Problem: Lessons from the Field
Jane StreetShriram Krishnamurthi, Emmanuel Schanzer, Kathi Fisler, Laurie, Askle, James, Shetha, Eric Alata, Kathy, Dharan, Kenny, Yaron, Alice O'Neill, Papert
- Sriram Krishnamurti plans to analyze curriculum design through problem-solving constraints and outline the Bootstrap program's three core goals: equity, rigor, and scale, with a specific focus on making these goals achievable in scalable settings rather than relying on non-scalable methods.
- Predictions regarding mandatory computer science education suggest that requiring it will likely cause scale to diminish due to the difficulty of finding qualified teachers for low pay, while also forcing a drop in rigor because everyone must pass to graduate.
- Conversely, making computer science an elective is predicted to reduce equity, as data indicates white and Asian males are heavily favored in optional classes, and students failing required courses like Algebra may have no room in their schedules for electives.
- After-school programs face significant risks regarding equity due to students needing to work or care for siblings, lack of rigor due to insufficient volunteer training time, and scalability issues in rural areas lacking corporate volunteer groups.
- Metrics for after-school programs indicate average volunteer return rates of 25 to 30 percent compared to 70 percent for Bootstrap, with students attending roughly 50 percent of sessions and often becoming exhausted by the end of the school day.
- The speaker anticipates that making computer science a universally applicable skill will allow it to be integrated into other subjects like economics or algebra, leveraging existing networks of teachers and organizations like the National Council for Teachers of Mathematics (NCTM).
- Challenges in integrating computing with mathematics include the fragility of skill transfer, deep structural differences such as functions not satisfying the vertical line test or integer division resulting in zero, and issues like overflow and underflow.
- To mitigate student frustration and "blank page syndrome," the Bootstrap approach utilizes a multi-stage design process via worksheets involving contracts, purpose statements, concrete examples, and abstraction, which has allowed students with no prior experience to build a video game in 20 hours.
- Control studies indicate measurable learning gains for students building video games, and the program currently includes algebra and data science curricula, with a new data science offering using the Pirate language (a parenthesis-free version of Racket) to teach queries rather than loops.
- Future plans include developing Bootstrap Junior for middle schools with a screen reader accessible and ARIA compliant block-based environment expected to be ready within six months to a year, alongside efforts to integrate pre-service teachers and MAT candidates into curricula.
- Strategic risks involve the danger that the chosen method for integrated computing is incorrect, the difficulty of obtaining longitudinal data due to privacy protections, and the challenge of scaling teacher training when only one teacher from a school participates.
- The Bootstrap organization, comprising about four people, predicts that while decentralization allows experiments to bloom, adoption is not guaranteed, and a major barrier to success is the isolation of individual math teachers without a team to train alongside them.