Information Technology Reference
In-Depth Information
Intelligence tests deal with discovering relationships and educing correlates by
noting similarities and differences among given facts. They are the measurement of
general intelligence, known as the 'g' factor in psychology. They assess our pattern
of abilities to discover the relevant qualities and relationships between objects or
ideas that are before us and to evoke other relevant ideas. No psychologists have
ever claimed that these tests can also measure other qualities such as character and
artistic talent. Controversial they may be, but most competent judges agree that they
provide a better measure of intelligence than any other at our disposal.
The validity of intelligence tests is also shown by the fact that their results are
highly correlated with other activities in which intelligence plays a dominant part.
For example, Eysenck noted the correlation between the results of intelligence tests
with the success of students at university (Eysenck
1974b
, p. 20). Students who
obtain a first-class degree have usually scored ten IQ points higher upon first entering
university than did students obtaining lower-class degrees; successful students have
usually scored some 15 points higher than students who failed to obtain a degree at all.
The close relationship found between IQ and the success at university is remarkable.
For the very same reason, intelligence tests have been profitably employed and
widely used by psychologists. They have also been used, for example, by school
medical officers in the diagnosis of mental deficiency (e.g. the Terman and Merrill
1936
,
1960
, New Stanford Revision of the Binet-Simon Scale was once the most
popular in Britain), by specialized training centers (such as pilot school), and by top
universities (Oxford and Cambridge) as a mean to assess their potential students.
Argument about the nature of intelligence has been going on since 1920s. Eysenck
pointed out in 1974 that “it is fair to say that it may even be that in the next twenty years
we will know a little more about the nature of intelligence than we do at present”.
He continued “until then we shall have to contend with our ability to measure it with
a certain degree of accuracy, and with such data as can be collected by means of
intelligence tests” (Eysenck
1974b
, p. 38). It is now 40 years later, and very little
progress has been made.
There is another reason the IQ test is being used. Throughout modern history, pro-
fessional psychologists devised intelligence tests. They depended on a pre-conceived
norm of a mental function, and were standardized so that their results would fall
within the normal or Gaussian curve of distribution. The standardization of an intel-
ligence test involves the expert use of statistical techniques. It also involves much
time and labor, because in order to establish norms, the tests were applied to a repre-
sentative sample of those for whom it is intended—a sample in which all the relevant
differences in the whole group are represented in their proper proportions—and the
distribution of scores for every age group were determined.
This standardization is very important in that it provides a scale to measure in-
telligence. In this respect, measurement of intelligence resembles measurement of
height—or, indeed, anything else. Knowing the height of a 16-year-old boy, we know
whether he is tall, short or average by comparing him with the average height of boys
of 16. Similarly, if we know the score the boy obtains in an intelligence test, we can
determine his brightness or dullness by comparing it with its appropriate norm. The
tests and the scale were rigorously tuned to reflect human intelligence throughout
the history of their construction in Europe and the USA.
