Intelligence and the differentiation hypothesis

I J Deary, V Egan, G J Gibson, E J Austin, C R Brand, T Kellaghan

Research output: Contribution to journalArticlepeer-review

Abstract

General intelligence (Spearman's g) accounts for over 50% of the reliable variance in a battery of mental tests in a sample of the general population. In a ''differentiation hypothesis'' originally suggested by Spearman it is hypothesized that the degree to which g pervades performance on mental tests is greater at lower ability levels. In addition to providing a critical review, the study presented here tests the differentiation hypothesis: (a) at different ability levels and ages; (b) when groups are selected on the basis of a wide range of criterion abilities; and (c) by developing new statistical techniques for sampling groups of different ability levels. Data used were the Differential Aptitude Test results of over 10,500 Irish schoolchildren aged 14 through 17 years. Of groups selected on the basis of verbal, numerical, or spatial ability the below-average ability groups had a more pervasive g factor, confirming the differentiation hypothesis.

Original languageEnglish
Pages (from-to)105-132
Number of pages28
JournalIntelligence
Volume23
Issue number2
Publication statusPublished - 1996

Keywords / Materials (for Non-textual outputs)

  • IQ LEVELS
  • HERITABILITY
  • ABILITY

Fingerprint

Dive into the research topics of 'Intelligence and the differentiation hypothesis'. Together they form a unique fingerprint.

Cite this