Students are more likely to retake the SAT if their score ends with '90'

The work was inspired by a study that found that a car's value drops suddenly when it passes a 10,000 mile mark -- so a car that has 70,000 miles is worth markedly less than one with 69,900 miles. "We were talking about that and we started thinking about SAT tests," says Uri Simonsohn of the University of Pennsylvania, who cowrote the study with Devin Pope of the University of Chicago.

Pope had a set of SAT scores from 1994 to 2001 -- before the SAT scoring system changed -- when the maximum score was still 1600. These scores were only the last score attained by each student, so if they retook the test, their first score didn't appear. The researchers found gaps just below 1000, 1100, 1200, and so on, indicating that people who got those scores were more likely to retake the test and have that just short of a "00" score replaced by something else.

The change in SAT scores probably doesn't make a big difference in the students' lives, Simonsohn says. "The SAT doesn't matter nearly as much for admission as people think, so 10 points probably don't make a difference." (In fact, when Simonsohn looked at actual admissions data, he found that students who scored 1390 were just as likely to be accepted as students who scored 1400.) His only worry is that students might be wasting their time retaking the SAT to reach a pointless goal rather than doing something more productive.

In experiments, the researchers also found that people who imagined running laps were more likely to say they'd do another lap if they'd just finished 19 than if they had already run 20. A look at baseball stats found that that players are four times more likely to end a season with a .300 batting average than a .299 average -- they manipulate their batting average by making decisions about whether to walk or swing, or whether to have a pinch hitter come in.

The research "tells you how important self-motivation is," Simonsohn says. People are surprisingly driven by round numbers and will take major action -- like sitting through a day of standardized testing, which hardly anybody enjoys -- to reach these arbitrary goals. Economists in particular tend to focus on actual awards that come from outside, like money or another reward, he says, but this is a clear example of motivation coming from within.