"Nationwide, the concept of 'CS for all' has become ubiquitous," says Brown CS Professor Shriram Krishnamurthi, co-director of Bootstrap. "But not all students will become computer scientists. We know already that opportunities to use data will be part of their lives ahead. To be successful, they need to seize those opportunities and use data well, and we want to help them do that." Bootstrap is a series of research-based K-12 curricula now used globally, and they've just launched a data science program with the goal of helping students learn to use data effectively.
"We wanted to provide a foundation for data science and statistical thinking," says former Brown CS undergraduate and current Master's student Sam Dooman, who co-taught the program at Providence, Rhode Island's Central High School. "In Bootstrap Data Science, students write programs to ask and answer authentic questions with data. By exposing students to these concepts in different contexts, we hope to provide more equitable computer science and statistics education."
Like other Bootstrap curricula, the data science program, which uses the Pyret programming language, integrates into existing classes, providing equitable access for school systems in which money isn't available to hire computer science specialists. Co-instructing with teachers Jennifer Geller and Tom Hoffman, Sam worked with a class of Social Studies students. For an inner-city school, the topic was extremely relevant: using Rhode Island Public School data to find correlations between school performance and demographics.
"The students wrote programs to answer questions that real sociologists ask," Sam says, "and the students were really motivated. Most had never programmed before, and they were all able to complete the course and answer questions about the dataset. There was one moment when a student who had no prior experience with programming helped a friend to debug his code, and we hadn't even taught debugging yet, which was really cool. They started the program afraid that they'd break the computers, and almost immediately overcame that fear. It was incredible to see how fast they got a handle on Pyret, getting error messages at the start and then working through them."
The Bootstrap team is now in editing mode, evaluating the success of Sam's work and perfecting the curriculum, adding more exercises and scaffolding worksheets, and making some improvements to Pyret. This summer, the program will begin again with a workshop in Colorado to prepare another cohort of teachers.
"Data science now drives many aspects of our lives," says Sam. "Whether it's with SQL tables or Excel spreadsheets, working with data is part of almost every profession. Teaching statistics is a very difficult task, and we've seen tremendous societal problems that could have been prevented with a better understanding of data science. We need students to be able to decompose statistical problems, look for structure, apply patterns, and answer questions with confidence."
And the new data science curriculum has been designed with exactly that in mind, providing an entry point with no prerequisites needed. "Our program is extremely accessible," Shriram explains. "It could be any kid's first exposure to computer science and/or statistical methods. For years, Bootstrap has been focused on teaching foundational skills, creating equity, and integrating into existing classrooms, and now we're tackling a longstanding problem of teaching statistics. It's a really comprehensive solution."
For more information, click the link that follows to contact Brown CS Communication Outreach Specialist Jesse C. Polhemus.