Skip to main content icon/video/no-internet

Introduction

There is an unlimited variety of human mental abilities, defined as any form of information processing capability that can be assessed objectively and quantitatively by means of psychometric tests or various laboratory apparatuses. Information processing includes diverse cognitive functions such as stimulus apprehension, attention, perception, sensory discrimination, generalization, conditioning, learning, short-term and long-term memory, recall, learning-set acquisition, concept formation, thinking, reasoning, inference, problem solving, planning, invention, and use of language. Quantitative assessments of such information processing functions by objective means typically show a wide distribution of individual differences. It is well established in psychometrics that individual differences in a wide variety of cognitive tasks, however diverse in specific knowledge content and required skills, are all positively correlated in the general population. This phenomenon of all-positive correlations among measures of individual differences in cognitive abilities is the basis of the theoretical construct of general ability, or g.

Factor Analysis of Mental Tests

The g factor is conceived technically as a latent variable that accounts for the empirical fact of all-positive correlations among diverse cognitive tests. By means of factor analysis one can determine the g factor loadings of various tests, i.e. their degree of correlation with the one factor that is common to a number of different cognitive tests, and from which g factor scores of individuals can be estimated.

‘General Ability’ and ‘Ability in General’

It is important to distinguish between general ability (or the g factor), on the one hand, and what can be called ability in general, on the other. The latter refers to the sum (or average) of the scores on a collection of different subtests, such as the Stanford-Binet and the Wechsler batteries, and many other heterogeneous tests of ‘intelligence’. The total score or Full Scale IQ on such tests is based on an arbitrary selection of a number of diverse tests. The g factor, however, in a linear decomposition of the total variance into uncorrelated components or factors, reflects only the source of variance that is common to all of the different ability measures represented by the various subtests of a cognitive test battery. Hence the simple sum of the subtest standardized scores on a test battery and the factor scores obtained from the g loadings of various subtests are not necessarily the same and may even be quite different. Typically, however, in the most widely used and broadly valid standardized test batteries the sum of the subtest scores (e.g. the full scale IQ) and the g factor scores are very highly correlated. Therefore the labour of calculating factor scores has little justification for the practical use of tests; the total standardized score, or IQ, is a fair substitute for the estimated g factor score.

Test Construction for the Measurement of g

The psychometric procedures for constructing an IQ test are intended to yield high practical validity for predicting many different outcomes involving cognitive abilities, such as scholastic achievement, college grade-point average, job performance, occupational level, and success in various armed services training programmes. This aim of IQ test construction and the psychometric means for achieving it inevitably results in the production of highly g loaded IQ scores, even when factor analysis is not used and the test designers have no interest in maximizing the test's overall g saturation per se.

...

  • Loading...
locked icon

Sign in to access this content

Get a 30 day FREE TRIAL

  • Watch videos from a variety of sources bringing classroom topics to life
  • Read modern, diverse business cases
  • Explore hundreds of books and reference titles

Sage Recommends

We found other relevant content for you on other Sage platforms.

Loading