Dressed in a pinstriped sports jacket and armed with a wheel trundle, two dozen stopwatches, and a wooden device used by the BC theater department to fake thunderclaps, US Science Chair Scott Rubin set out for Prospect Park on the afternoon of September 13, 2012 with his Upper School Physics class. The sports jacket was a holdover from Mr. Rubin's years teaching in South Africa ("When I came to Berkeley Carroll, I realized that only Mr. Swarthout was dressed this way," he says, "but I was three days into classes and thought I might as well keep it up").

The other items were for measuring the speed of sound. It's 343.2 meters per second, of course—but as Mr. Rubin explained, the lab was actually intended “to get students to think about uncertainty, about what can cause our measurements to go wrong.”

Pairs of students stood 100 meters apart in the park as Mr. Rubin—standing with half the kids—clapped the wood blocks. Students started their stopwatches when they heard the thunderclap, then switched places with their partners and stopped their watches when Mr. Rubin fired off another thunderclap. Partners compared the difference between their times, repeated the experiment, and compared results between groups. In class the next day, they discussed the difference between random and systematic error, and various methods they could have employed in the lab to reduce uncertainty (from taking multiple trials to measuring larger quantities).