Correction of Flynn Effect

...

Discussions:

First of all, we can see the Flynn effect we got from this meta-analysis is 2.31, which is lower than Flynn’s (3.11) and Fletcher’s (2.80). When we calculated the modern set meta-analysis, the result would be 2.93 (with a standard error is 0.35), this is the most reasonable estimate of Flynn effect since 1972.

For ability level, the mean magnitude may make no sense for changing significantly with level of ability, and this correction can be used to scores across ability level.

For age, Flynn effect has no difference between different age groups.

For sample type, sample type effect was not statistically significant, but we may increase Flynn effect as control over these variables is relaxed.

For order of test, when the newer test is ordered first, the excepted Flynn effect is about 0.20-0.29, while when the older test is ordered first, the excepted Flynn effect is about 0.20-0.35.

For pairing, we can say similar corrections can be made to one of the Wechsler and Binet tests since 1972.

We will see the condition about kinds of hypotheses below:

Genetic hypotheses, the larger Flynn estimate in our study for newer than older tests provides no compelling support for the heterosis hypothesis.

Environmental factor, Flynn effect does not diminish over time and even may be larger for modern tests. The effect was similar across age and ability level. We may anticipate the difference in IQ score based on family factors, educations as so on.

For measurement issues, we should note caution in estimating the degree of norms obsolescence for the KABC and different screening tests.

Assessment: The results admit a significant and continuous elevation of IQ test norms as described by Flynn. We have to see the importance of correcting for the Flynn effect in high stakes decisions.

Conclusion:

There is no doubt that the need to correct high stakes IQ test scores for norms obsolescence decision-making is abundantly clear. In death penalty cases, a score difference of 95 and 98 make no sense, but for 71 and 68, or 76 and 73, it may be a criterion for death and life. When we consider the influence of time, we could use the same test over time. It means we need to adjust the cut point of intellectually disability.

We never deny there are unknown factors influences the Flynn effect, but we should not omit our main findings during this study: the Flynn effect for modern set is around 3 per ten years, most distribution was greater than 0. These findings are consistent with the previous studies, so it should be advisable to correct IQ scores for Flynn effect.

References:

The Flynn Effect: A Meta-analysis Lisa Trahan, Karla K. Stuebing, Merril K. Hiscock, and Jack M. Fletcher University of Houston Published in final edited form as: Psychol Bull. 2014 September ; 140(5): 1332–1360. doi:10.1037/a0037173.

Agbayani KA, Hiscock M. Age-related change in Wechsler IQ norms after adjustment for the Flynn

effect: Estimates from three computational models. Journal of Clinical and Experimental

Neuropsychology. 2013; 35(6):642–654. [PubMed: 23767697]

American Association on Intellectual and Developmental Disabilities. Intellectual disability:

Definition, classification, and systems of supports. Washington DC: American Association on

Intellectual and Developmental Disabilities; 2010.

American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, fourth

edition, text revision (DSM-IV-TR). Washington, DC: American Psychiatric Association; 2000.