Garcia and Tor first compared SAT scores for locations that had a lot of people in the room taking the test versus locations that had smaller numbers. They adjusted the scores to control for the educational budget in that region and other factors. Students who took the SAT test in a room with less people scored higher.