Spearman’s Hypothesis Tested Comparing 47 Regions of Japan Using a Sample of 18 Million Children

Kenya Kura,Edward Dutton,Jan Te Nijenhuis

doi:10.3390/psychology1010002

Kenya Kura, Edward Dutton + Show 1 more

Open Access

https://doi.org/10.3390/psychology1010002

Copy DOI

Abstract

Many groups differ in their mean intelligence score. Spearman’s hypothesis states that the differences are a function of cognitive complexity. There tend to be large differences on subtests of high cognitive complexity and small differences on subtests of low cognitive complexity. Spearman’s hypothesis has been supported by a large number of studies. Can Spearman’s hypothesis be generalized to regions of a country, where these regions differ in mean intelligence? We utilized data from 86 different cognitive tests from all 47 Japanese prefectures and correlated the g loadings of 86 subtests with standardized differences on the same subtests. Spearman’s hypothesis was clearly supported: the biggest differences between the regions were on the tests that were of the greatest complexity, meaning that Spearman’s hypothesis may be generalizable from groups to regions. In Japan, g loadings offer a better explanation of group differences in intelligence than cultural differences. Future research should explore whether Spearman’s hypothesis is also supported for differences between regions of other countries.

Highlights

Charles Spearman was the first to postulate that there exists a single general factor of human intelligence, something that reflects positive correlations among different cognitive tasks
Spearman’s hypothesis was clearly supported: the biggest differences between the regions were on the tests that were of the greatest complexity, meaning that Spearman’s hypothesis may be generalizable from groups to regions
In Japan, g loadings offer a better explanation of group differences in intelligence than cultural differences

Summary

Introduction

Charles Spearman was the first to postulate that there exists a single general factor of human intelligence, something that reflects positive correlations among different cognitive tasks. Spearman coined the term “g factor” [1] He hypothesized g as something like mental energy which enables various cognitive activities, such as memory, deduction, induction, grasping abstract relationships, rule inference, and finding similarities and dissimilarities. Following this lead, Jensen maintained that g should be treated as a distilled entity obtainable by factor analysis from all mental tasks because, argued Jensen, it stems from individuals’ neural and physiological substrates [2]. It would be hard to imagine that an artificial characteristic limited to intelligence tests and their construction can be somehow correlated with so many human variables

Methods

Results

Conclusion