Statisticians are one of the fastest growing jobs in the country–and for good reason. “As data becomes more and more a part of people’s lives, statistics is becoming more a part of people’s lives,” says Daniel Kunin, a Brown University student who’s created an interactive textbook for introductory statistics classes in high school and college. “It is a dense topic that a lot of people want to learn more about and either haven’t had the time or don’t have access to formal education.”
Kunin’s interactive online resource is called Seeing Theory, and the website features five areas of statistics, including basic and compound probability, distributions, statistical inference, and linear regressions. Each of these topic areas is broken down into a series of three interactive graphs that illustrate concepts while letting users play around with data. The idea is to make statistics more tangible and accessible to people who might not grok the concepts immediately.
“As the number of students and the backgrounds of students that are taking statistics grows, I think the pedagogy around statistics has to adapt too,” Kunin says. “There are a lot of students taking it with less math literacy, but have to to take it for other disciplines. A visual approach might be the link for those students.”
Kunin is a senior at Brown, where he’s studying applied math and computational biology. He has mostly taken classes in math and computer science, but in the fall of 2015, he took a course on data visualization and information design. For his final project, Kunin decided to visualize statistical concepts, and Seeing Theory was born. After the semester was over, Kunin wanted to continue the project, and applied for grants from Brown and the National Science Foundation group STATS4STEM which enabled him to continue the work over the next two years. He also worked with a RISD student, Jingru Guo, on the structure and graphic design elements of the site.
“There’s a lot of movement inherent in the mathematics that lends itself really well to interactive visualizations and not as well to some of the diagrams you might see in a classic textbook,” Kunin says, referring to how statistics often relies on taking samples from a population over time and looking for how the data converges into trends–all of which lend themselves to illustration. He hopes that Seeing Theory will be used primarily as a teaching tool–it’s not meant as a quick guide to statistics that you can absorb in an hour. But when students turn to Google for help instead of their textbooks, he hopes they’ll stumble across Seeing Theory.
Seeing Theory isn’t done yet, but Kunin hopes to finish it up by the summer as well as incorporate some of the criticisms he’s received, particularly that some of the descriptions of the concepts aren’t as clear as they could be. He also would like to translate the site into other languages to make it more widely accessible.
After all, the more data literacy, the less data bullshit.