Intelligence test meta-analysis supports Flynn Effect
Posted on 16/01/2015
The Flynn effect predicts that scores on intelligence tests will increase year-on-year and this effect, first coined in 1994, has been supported by a new meta-analysis from researchers at the Institute of Psychiatry, Psychology & Neuroscience (IoPPN) that included scores from over 200,000 participants.
Peera Wongupparaj, a research student in the Department of Psychology, investigated changing IQ scores in 48 countries across the world and over a 64 year time period and concluded that the Flynn effect holds up in each of the five age groupings used as well as in different nations. In low income countries, the Flynn effect was observed to a greater degree than in high income countries over the period and although the reasons for this larger increase are not clearly established, potential causes include improved education, medical care and nutrition, modernised child rearing, increased exposure to testing and even the use of artificial lighting. As a result of the accelerated Flynn effect in low income countries, the gap between the average IQ in low income and high income countries has narrowed and now stands at just below three points.
“Defining the different type of countries is an issue for this type of study and so we based high/low income countries on recent International Statistical Institute criteria that considered on their Gross National Income (GNI) per capita per year,” said Peera Wongupparaj. “Accordingly, we were able to include in the meta-analysis 532 studies from high income and 198 from low income countries.”
The Flynn effect is named after James Flynn, Emeritus Professor of Political Studies at the University of Otago, New Zealand and was coined by Richard Herrnstein and Charles Murray in their controversial 1994 book The Bell Curve. The concept of the Flynn effect is that performance on IQ tests improves over time in new sample populations. When an IQ test is originally calibrated the Intellectual Quotient (IQ) is set such that 100 represents an average score. In part, because of the Flynn effect, this calibration has to be done again every so often to take into account the changing performances. The work by Peera Wongupparaj, supported by his supervisors at the IoPPN Professors Robin Morris and Veena Kumari, has made a major contribution in confirming this effect using newly applied statistical techniques, and a key conclusive finding that it is the accelerated Flynn effect in low income countries.
Peera Wongupparaj said: “We used a novel statistical technique in this context called Cross-Temporal Meta-Analysis, which allows for the bias in such studies caused by varying sample sizes and within-study variance, and applied this technique to Raven’s Progressive Matrices IQ data taken from published articles during 1950 to 2014.”
Professor Robin Morris said: “This study clearly now confirms the Flynn effect beyond reasonable doubt and the manner in which intellectual test performance is catching up in low income countries. The challenge, of course, is to define more clearly why the increases have been taking place.”
Paper reference: Wongupparaj, P, Kumari, V and Morris, R G ‘A Cross-Temporal Meta-Analysis of Raven's Progressive Matrices: Age groups and developing versus developed countries’ published in Intelligence DOI: 10.1016/j.intell.2014.11.008
For further information contact Tom Bragg, Press Officer at IoPPN, King’s College London, on +44(0)2078485377 or email email@example.com