Dec. 2, 2003

The year in which an IQ test is given can make the difference between life or death, Cornell researchers find

ITHACA, N.Y. -- The year in which IQ is tested can make the difference between life and death for a death row inmate. It also can determine the eligibility of children for special services, adults' Social Security benefits and recruits' suitability for certain military careers, according to a new study by Cornell University researchers.

That's because IQ scores tend to rise 5 to 25 points in a single generation. This so-called "Flynn effect" is corrected by toughening up the test every 15 to 20 years to reset the mean score to 100. A score from a test taken at the end of one cycle can vary widely from a score derived from a test taken at the beginning of the next cycle, when the test is more difficult, says Stephen J. Ceci, professor of human development at Cornell.

Ceci and his current and former graduate students, Tomoe Kanaya and Matthew Scullin, found, for example, that the number of children recommended for special services for mild mental retardation tripled during the first five years of a new test compared with the final five years of an old test, despite the fact that there were no real changes in underlying intelligence.

"Our findings imply that some borderline death row inmates or capital murder defendants who were not classified as mentally retarded in childhood because they took an older version of an IQ test might have qualified as retarded if they had taken a more recent test," Ceci says. "That's the difference between being sentenced to life imprisonment versus lethal injection."

The study is published in the October issue of American Psychologist ( Vol. 58, No. 10, pp. 778-790), a journal of the American Psychological Association. Co-author Kanaya is a fourth-year graduate student in human development and is the first author. Scullin, Cornell Ph.D. '01, is now an assistant professor of psychology at West Virginia University and is the second author.

The researchers analyzed IQ data from almost 9,000 school psychologist special education assessments in nine school districts across the country to document how the resetting of the IQ test influences mental retardation diagnoses for several years after a new test is introduced.

The consequences of taking intelligence tests at the end or beginning of a test's cycle are most critical, however, when determining whether a death row inmate is mentally competent. Of the 350 people executed since 1990, 112 were known to have IQ scores of 70 or below (the cutoff for mental retardation).

Among children, the researchers found nearly a six-point difference between those taking the two tests. "This variance can make the difference between a child being diagnosed as mentally retarded or not," Ceci says. "This study shows for the first time that two children in the same classroom with the same cognitive ability could be diagnosed differently simply because different test norms were used for each child."

The researchers report that perhaps tens of thousands of children could be affected by these IQ trends over the course of their school years, with far-reaching financial implications. "Our results imply that millions of taxpayers' educational dollars may be misallocated because students are being misdiagnosed every year that an IQ test ages," Ceci points out.

A diagnosis of mental retardation also determines whether a person is eligible for Social Security disability benefits. And the year in which a military recruit takes an IQ test can determine whether he or she is eligible for service or certain occupations and ranks.

"Caution must be used when IQ scores are used to base important financial, social or legal decisions. It may not be sufficient to simply look to see if an IQ score is below some cutoff point," concludes Ceci. "The most important times to be particularly careful are when the test is either at the beginning or the end of its cycle."

The research was supported, in part, by a grant from the Smith Richardson Foundation to Ceci.

Related World Wide Web sites: The following sites provide additional information on this news release. Some might not be part of the Cornell University community, and Cornell has no control over their content or availability.

oFull text of the article is available from the APA Public Affairs Office or at:

oStephen Ceci: