Competitive Timed Tests Might Be Contributing to the Gender Gap in Math

By Emily Richmond
banner_AP girls math.jpg
Third-grade gifted and talented math class students Molly Trevor, 8, Matt Stec, 8, and Austyn Long, 8, learn to use the memory function on the calculator. (Brian Branch-Price/AP Images)

When I was in the fifth grade, Fridays always meant math drills. Our teacher stood over us with a stopwatch in one hand and a gym whistle in the other. At the first piercing blast, we would rush to solve as many of the equations as we could before time was up. Our names and rankings (a rainbow scale starting at purple and progressing up to red) were posted prominently on the wall, and the top achievers won prizes.

It didn't take me long to fear the whistle. After most of my classmates—particularly the boys, as I recall—had advanced up to yellow and even orange, I lingered in the lowly blues. I'm not sure how much of this experience factored into my later struggles with math, but it probably didn't help.

Now a new study, published in the Journal of Economic Behavior and Organization, suggests that relatively simple twists on the familiar high-speed, high-pressure math quiz can level the field of classroom competition.

Research has long shown that males respond better to competitive incentives than their female counterparts. And there are plenty of studies that have found when boys and girls are put in head-to-head competition in which there's a single, timed opportunity to win, boys excel. For the new study, researchers explored what happens when students are given a second chance to compete, and how eliminating the time limits further changes the outcomes.


Educated Reporter logo
Students, Teachers Push Back Against High-Stakes Testing

Girls and STEM Education: Still Waiting For Liftoff

Is Student Motivation the Missing Link in School Reform?

The experiment was conducted with more than 500 students at 24 Utah elementary schools. Boys and girls, each competing against a classmate, had to answer as many math questions as they could in a five-minute period. Prizes went to the winners.

In the first round of the timed competition, boys performed better than girls, reinforcing the conclusions of earlier studies. But when a second round was added, the advantage for boys disappeared. Here's where it gets even more interesting: When the competition was extended to three rounds, girls began to outscore boys. And the first-round advantage for boys disappeared if the time element was removed from that competition.

"One of the reasons girls don't do well in competitive settings is that they don't think they're as good as boys—but they really are," said Brigham Young University economist Joseph Price, one of the study's co-authors. "That's an information problem, rather than evidence that girls are destined for a certain outcome." (You can read my full Q&A with Price at EdMedia Commons.)

That "information problem" could also have another component, known to researchers as "stereotype threat." Put simply, if people are worried about confirming negative perceptions of a group of which they are a member, it can hurt their individual performance. The seminal study on stereotype threat was published in 1995 by Claude Steele (now dean of the Stanford Graduate School of Education) and Joshua Aronson (associate professor of applied psychology at NYU), and found that black college students fared worse when they were asked to identify their race prior to taking a high-stakes exam. Numerous other researchers have since found similar results when asking people to identify themselves by a group—be it ethnicity, gender, socioeconomic status, or age, just to list a few—and then testing them in a subject where there are stereotypes about their perceived inferior ability.

Jill Pipher, a professor at Brown University and past president of the Association for Women in Mathematics, said she was surprised by some of the new study's findings, particularly that the lack of advantage for boys persisted when the experiment was re-run two weeks later.

"It does suggest that there is a lack of confidence among girls that disappears after exposure and acculturation," Pipher said. She noted that research has found that when girls were asked to identify their gender prior to taking the Advanced Placement math exam, they scored lower than they did when the question wasn't asked until after the test—or not at all. (That field study, conducted by the Education Testing Service which administers the AP exams, also found asking the gender question ahead of the exam boosted performance among boys.)

And just how much of learning mathematics is about getting something right in one shot, working within a rigid time limit? Isn't it more about mastering concepts and building skills over a longer time frame, and having the patience to tackle challenging problem sets that might require multiple attempts?

Linda Gojak, president of the National Council of Teachers of Mathematics, told me that she wasn't aware of many situations outside of the classroom where a "competitive, timed environment" was used for math.

"Too often the focus is on getting the right answer and getting it quickly—the process and thinking used to get an answer is just as important," says Gojak, director of the Center for Mathematics and Science Education, Teaching and Technology at Ohio's John Carroll University. "How often do you need to solve the exact same problem twice? The ability to adapt the reasoning used to solve one problem to a new situation is a key characteristic of students who are proficient in mathematics."

While the Utah study still needs to be replicated, it does pose some interesting questions for educators to consider—particularly those who are busy looking to explain and respond to differences in the ways boys and girls perform in school. But it's also important not to generalize about either group, said Lise Eliot, a neuroscientist and the author of the book "Pink Brain, Blue Brain," in which she builds a case that gender differences in children are influenced heavily by social factors and that it's shortsighted and potentially harmful to attribute them to biological hard-wiring.

"There's a lot of individual difference in competitiveness among boys and girls; it's not an absolute gender difference," Eliot told me. "It's yet another dimension that teachers need to be sensitive to. As far as competition as a motivator, I don't think we should cut it out of the classroom entirely."

The challenge for teachers, Eliot said, is to offer multiple opportunities and venues for students both to test their skills and have opportunities to excel, balancing the competitive with the collaborative. "Both of these are key skills for later in life," Eliot said. "You have to compete to get the job, but later you have to be able to collaborate with your co-workers."

Economist Joseph Price shared a similar sentiment, saying he hoped his study would encourage educators to see competition "as one of a large set of tools used to encourage student effort." He's in the middle of expanding his study to more than 30 elementary, middle and high schools in Utah. In the meantime, he's hopeful that teachers will let go of the notion that competition in and of itself is a deterrent to girls' learning. "Any concern (teachers) might have that competition will put girls at a disadvantage, that concern should be short-lived," Price told me. "Once girls acclimate to these situations they should do very well."

It's tough for me to judge the effectiveness of those high-speed, high-pressure math drills back in elementary school. But I do have to wonder how my performance, and the performance of other learners like me, might have been different if my teacher had put down the stopwatch—or at the very least, dropped that dreaded whistle.


This post also appears at The Educated Reporter, an Atlantic partner site.

This article available online at:

http://www.theatlantic.com/sexes/archive/2013/03/competitive-timed-tests-might-be-contributing-to-the-gender-gap-in-math/274406/