Renorming IQ Tests Due To Flynn Effect May Have Unintended Consequences

Researchers Tomoe Kanaya, M.A. and Stephen J. Ceci, Ph.D., of Cornell University and Matthew H. Scullin, Ph.D., of West Virginia University used IQ data from nearly 9,000 school psychologist special education assessments from nine school districts across the U.S. to document how the Flynn effect influences mental retardation diagnoses for several years after a new test is introduced. The students (ages 6 – 17) were from different geographical regions, neighborhood types and socioeconomic status.

The results show that the test renorming due to the Flynn effect influences which children are diagnosed with mental retardation regardless of their actual cognitive ability. According to the researchers, children in the same classroom with the same cognitive ability could be diagnosed differently simply because different test norms were used for each child. Students in the borderline and mild mental retardation range lost an average of 5.6 IQ points when retested on a renormed test and were more likely to be classified mentally retarded compared with peers retested on the same test, according to the study.

Specifically, when a commonly used IQ test (the Wechsler Intelligence Scale for Children or WISC) was renormed to account for the Flynn effect, the number of children in the borderline mental retardation range (IQ 66-70) recommended for special school programs due to mild mental retardation tripled during the first five years of the new test compared to an equivalent IQ range during the last five years of the previous test.

"Some students who would be eligible for mental retardation services under new IQ norms will fail to receive them because the older norms of the IQ test they were given allowed them to score above the cutoff," write the researchers. "Also, students who would not have qualified for mental retardation services had they been tested a year earlier will now do so if they are given an IQ test with newer, harder norms." Not only does this deny services to some students who need them, it wreaks havoc with school system's special education budgets because there are dramatic changes in the numbers of children that will qualify for special educational services.

Besides educational and financial issues, the Flynn effect and the renormed IQ tests can also have important legal implications. Nowhere are the consequences of IQ score fluctuations due to the Flynn effect more critical, say the authors, than in the determination of whether a death row inmate can be considered mentally retarded. "Our results imply that the year that a capital murder defendant was tested can determine whether she or he is sentenced to die as opposed to life imprisonment. This raises concerns regarding inmates on death row who tested above the 70-75 IQ cutoff on a test that was near the end of its norming cycle – when scores are highly inflated as well as an inmate who tested in the mental retardation range during the earliest years of a new norm – when the test is hardest," according to the researchers.

The Flynn effect also has military occupational consequences, according to the study. Depending on which IQ norms are used, a military recruit might be eligible or not for military service and influence whether they will be allowed to enter certain occupations or occupy certain ranks in the military.

"The main conclusion that can be drawn from these results is that caution should be used when basing important financial, social or legal decision on IQ scores," say the researchers. "Perhaps the most important times to be particularly cautious are when a test is either at the beginning or at the tail end of its norming cycle. Although test scores are most valid at the beginning of a norming cycle, they run the greatest risk of being compared to highly inflated scores from the waning years of the previous norming cycle."

Article: "The Flynn Effect and U.S. Policies. The Impact of Rising IQ Scores on American Society Via Mental Retardation Diagnoses," Tomoe Kanaya, Cornell University, Matthew H. Scullin, West Virginia University and Stephen J. Ceci, Cornell University; American Psychologist, Vol. 58, No. 10.