Test scores might not be telling us the real story about kids
Mar 11, 2013, 2:18 PM | Updated: 2:18 pm
Test-taking as an accountability measure is rampant in U.S. schools, and the results can have a big impact on how teachers and schools are evaluated. But a pile of research shows that test scores are subject to subtle manipulation by a variety of factors outside the test-takers' conscious control.
The belief that competition boosts performance isn't necessarily true when it comes to test-taking, for instance. That's according to “Top Dog,” a new book about the science of winning and losing by Po Bronson and Ashley Merryman.
The website for the book says it “reveals hidden factors behind every sort of win and loss —from bringing home a Gold in Olympic swimming to bombing the SAT.”
As for the latter, a review in Time magazine said the book describes an experiment with 124 Princeton underclassmen made up of questions from a graduate school admissions test. The test was labeled as an “Intellectual Ability Questionnaire, and included explanations meant to make it seem as threatening as possible. A second set of students took the same test, but with the less-threatening name of “Intellectual Challenge Questionnaire.”
The intimidated students in the first group got 72 percent of the questions right, and their more relaxed peers got 90 percent right, Time said.
Social factors have a big influence over intelligence, a fact that plays out when test-taking time rolls around, according to a story in The New York Times.
“Stereotype threat” is a problem for members of groups worried about confirming negative stereotypes about their group, the story said — including African-American and Latino college students, and females enrolled in math and science courses. Members of these groups got much lower scores on tests when their gender or race was brought up before the tests.
White male engineering students fell victim to testing voodoo, too — when reminded that Asians outperform other students on tests of math ability, the New York Times said.
Traumatic events in students' lives also make a difference in test-taking. A study conducted in a dangerous area of Chicago showed that students' IQ scores fell by a measurable amount after a murder occurred in their neighborhoods.
“This research has important implications for the way we educate our children,” the New York Times story said. “For one thing, we should replace high-stakes, one-shot tests with the kind of unobtrusive and ongoing assessments that give teachers and parents a more accurate sense of children’s true abilities. We should also put in place techniques for reducing anxiety and building self-confidence that take advantage of our social natures.”