The continued year-on-year rise of IQ test scores world wide, but with varying rates across countries. From 1952 to 1982, IQ rose on average by 20 points, giving a mean rate of increase of about 3 points per decade. The highest gains were for so-called culturally-reduced tests measuring fluid intelligence (e.g., Raven’s Progressive Matrices), and lowest for those measuring verbal abilities. The increase was highest in Belgium, Holland and Israel (20 points per generation), and lowest in Denmark and Sweden (10 points per generation). This cross-generational trend remained unnoticed due to the fact that IQ scores are calculated relative to the average score for the contemporary cohort, with the average being set to 100. If an IQ of 120 is compared with the average for a cohort tested one generation earlier, the score would then be about 130. Consequently, IQ tests require frequent renormalization. The effect is named after James R. Flynn who first published the corroborative data mentioned above in 1984 and then again in 1987. There is no straightforward explanation for the effect, but Flynn himself favors an increase in some sort of abstract problem-solving ability rather than a rise in intelligence per se.
See Cohort, Cohort effect, Cross-sectional design, Crystallized intelligence, Décalage, Equivalence (of data across cultures), Fluid intelligence, Intelligence, Problem solving, Raven’s Progressive Matrices