You are using an outdated browser.
Please upgrade your browser
and improve your visit to our site.

IQ Scores Shouldn't Be A Matter of Life or Death

For most of us, IQ tests are little more than another form of online procrastination. We put far more stock in GPAs and other qualifications. Yet until Tuesday, the difference of a single IQ point could be a matter of life or death—at least for inmates on Florida’s death row.

In a 5-4 ruling, the Supreme Court overturned a law that used a strict IQ cutoff point to determine whether an inmate was eligible for capital punishment. The case hinged on 68-year-old Freddie Lee Hall, who is accused of murdering a pregnant 21-year-old woman in 1978, and whose IQ scores straddle the threshold for mental disability. In 2002, the Supreme Court prohibited capital punishment for inmates deemed mentally retarded; in keeping with guidelines in the Diagnostic and Statistical Manual of Mental Disorders, the state of Florida has traditionally used an IQ of less than 70 as an indicator of intellectual disability. In tests taken between 1968 and 2008, Hall’s scores have ranged from 60 to 80, with his most recent scores fluctuating between the high 60s and the low 70s.

The intellectual problems inherent in trying to reduce a phenomenon as complex as intelligence to a single number—and a history of intelligence tests being used to justify racism—have made IQ tests unpopular among the politically correct. But there are methodological problems with giving IQ scores so much weight, too. People’s scores change over the course of their lifetime: Most people’s scores begin to decline in their 50s or 60s. Different types of tests can also yield different results.

“The concept that people are born with a certain level of intelligence and it remains the same throughout their life is not true," said Robert Sternberg, a Cornell professor who has served as president of the American Psychological Association. "IQ varies with the kind of parenting you have, the kind of schooling you have.”

“For scores close to cutoffs, one can ascribe little meaning since they are well within the standard error of measurement,” said Sternberg. And determining what’s “normal” or “deficient” changes based on what population you’re comparing an individual to—“To people in this country, in the world, in the South? There are differences across groups,” said Sternberg.

Comparing scores obtained over decades raises even more complications. Since the beginning of the twentieth century, IQ scores have been rising at a rate of about three points every ten years. “It’s really a mess,” said Sternberg. “There’s this totally false pretense that an IQ of 70 is the same in different time periods, but the conversion of raw scores isn’t stable over time.”

Plus, inmates have a pretty strong incentive not to do their best, and even psychologists who specialize in intelligence testing have trouble detecting deception. As Sternberg put it, “Why in the world would anyone want to do their best on a test that might result in their being executed?” It doesn’t take a high IQ—or any other measure of intelligence—to figure out that a prison IQ test isn’t the time to show off.