This is an overview of key players in the modern history of intelligence testing – many of whom you may
already be familiar with from their work on other aspects of psychology.
1. Francis Galton:
The 1st Modern Attempt (late 1800s)
Intelligence tests are grounded in the
work of Francis Galton in the late 19th century. Galton is considered to
be the father of the study of individual differences [and, incidentally
was the half-cousin of Charles Darwin]. For Galton, measurement of
intelligence was to be as direct as possible a measure of underlying
intelligence. Hence, Galton suggested reaction time as a feasible
approach and pursued various sensori-motor measurements.
Interestingly, more contemporary "Galton-like" approaches are being
pursued by psychologists such as Arthur Jensen and Mike Andersen who
assert that they are assessing the integrity of the central nervous
system - inherent capabilities (Intelligence A) - as opposed to
manifestations of intelligence in everyday life (Intelligence B).
2. Alfred Binet:
The origins of IQ Testing (early 1900’s)
Alfred Binet is celebrated in history
as the man who created the first 'intelligence test' in the form as we
know them today. He is commonly known as the “father” of IQ testing.
In 1904, Binet was commissioned by the French Ministry of Public
Instruction to develop techniques for identifying primary grade children
whose lack of success in normal classrooms suggested the need for some
form of special education (Gould, 1981).
In 1905 he produced the Binet-Simon scale [with Theodore Simon] - the
first intelligence test. Binet took a pragmatic approach, choosing a
series of 30 short tasks related to everyday problems of life (e.g.
attend to simple instructions
- name parts of the body
- compare lengths and weights
- counting coins,
- assessing which of several faces is 'prettier',
- naming objects in a picture,
- digit span (the number of digits a person can recall after being shown
a long list),
- word definition
- filling in the missing words in sentences, etc.
Supposedly all these tasks involved basic processes of reasoning.
The tests were arranged so as to be of increasing difficulty. Each level
of tests matched a specific developmental level - i.e. all tests at a
given level were capable of being solved by any normal child in that
This was a turning point in psychology: A new type of test had been
produced in which the average level of performance was the criterion. In
1908 the test was revised and then again in 1911. This edition was the
model for many future tests. The test results proved to be correlated
with other criteria (e.g. results of school examinations, assessments of
4. Terman (1916) and Stern (1912)
Lewis Terman (1877-1956) of Stanford
University in the US decided to use Binet's test. He found that the
Paris-developed age norms didn't work very well for Californian school
children. So he revised the test: adapted some items, added other items,
established new age norms, and extended the upper age limit to "superior
adults". This became the Stanford-Binet revision in 1916. In this
revision the Intelligence quotient first appeared. The Intelligence
Quotient or IQ was a score meant to quantify intellectual functioning to
allow comparison among individuals. To arrive at an IQ score, Terman
relied on a formula expressing the relation between an individual’s
mental age and chronological age developed in 1912 in Germany by
Stern: 1912 Wilhelm Stern proposed the following formula:
IQ = mental age x 100
This formula works fairly well for children but not for adults (Thomson,
1968; Weinberg, 1989)
5. Yerkes: Army
The US army at the beginning of WWI was
faced with the problem of assessing the intelligence of great numbers of
recruits in order to screen, classify, and assign them to suitable
tasks. The Stanford-Binet test required a highly trained person for
individual administration - thus it would prove time consuming and
costly for large-scale use.
So, when the US entered WWI in 1917 a committee was appointed by the APA
to consider ways that psychology might assist the conduct of the war.
Head of this committee was Robert Yerkes. His brief was to develop group
Yerkes, a psychologist and army major, assembled a staff of 40
psychologists [including Terman] to develop a group intelligence test.
This resulted in the Army Alpha and Army Beta tests. The Beta was a
version of the Alpha specifically for use with non-English-speaking and
illiterate persons. [Instructions to those taking the Beta were given by
demonstration or pantomime, rather than orally or in writing.]
In the end, over a million people were tested, but not until late in the
war. Thus the work actually had little effect on the war, but did a
great deal to enhance the status of psychology. After the war, industry,
business and education saw potential value of psychological testing
Note that the validity of the Beta test, in particular, has since been
questioned rather damningly by Gould (1981).
Spearman: “g” (1920’s)
Up to now, the approaches to
intelligence had been very pragmatic - i.e. tests were developed for
particular needs. However, another approach to understanding
intelligence, involved analysing data that was already collected.
Charles Spearman (1927) analysed the relations among experimental
intelligence tests using 'factor analysis'. He argued that, as a rule,
people who do well on some intelligence tests also do well on a variety
of intellectual tasks [vocabulary and mathematical and spatial
abilities]. And if people did poorly on an intelligence test, then they
also tended to do poorly on other intellectual tests. That is, he
observed correlations among performance on a variety of intellectual
Thus, he proposed, a 'two-factor' theory of intelligence:
- General Ability (g): which was required for performance of mental
tests of all kinds; he called this a kind of 'mental energy' that
underlies the specific factors
- Special Abilities: which were required for performance on just one
kind of mental test.
- e.g. Scores on a verbal comprehension test are largely determined by
one’s level of general intelligence but they are also affected by one’s
specific ability to perform verbal comprehension tasks.
But the main thrust of Spearman's analysis was this idea of a general
intellectual capacity. This formed a major theoretical platform for many
subsequent approaches to intelligence.
It might be also noted, however, that Spearman was perhaps excessively
enthusiastic about g. For example, he advocated restricting voting
rights to people whose g exceeded a certain level, and he was a
eugenicist (eugenics comes from the Greek "eugenes" meaning well-born) -
arguing that only people with a certain level of g should be allowed to
"g" was controversial then as now.
Intelligence Scales for Adults and Children (1939 - present)
Another test designer was
Wechsler. Wechsler felt that the Binet scales were too verbally loaded
for use with adults, so he designed an instrument with sub-tests to
measure both verbal and nonverbal abilities, largely borrowing from many
other tests, such as the US Army Alpha test. He adopted a mean score of
100, since the Stanford-Binet metric had become universally accepted.
The original Weschsler-Bellevue test in 1939 proved quite successful in
civilian and military applications.
In 1949, Wechsler produced the Wechsler Intelligence Scale for Children
(WISC), which competed with the Stanford-Binet test.
In 1955, he produced a revision of the adult scales named the Wechsler
Adult Intelligence Scale (WAIS). And later he produced a scale which
could be used with pre-primary children.
These scales have all been revised, but still show a distinct
resemblance to the original 1939 scale.
Primary Mental Abilities (1930’s)
Another 'factor analyst',
(1938), accepted Spearman's hypothesis of a general factor. But he
disputed its importance. He argued that g is in fact a second order
factor or phenomenon - one which arises only because the primary or
'first-order' factors are related to one another. Thus, Thurstone
identified 7 'primary mental abilities' which he judged to be more
important. These were:
1.Verbal Comprehension: vocabulary, reading, comprehension, verbal
2. Word fluency: the ability to quickly generate and manipulate a large
number of words with specific characteristics, as in anagrams or rhyming
3. Number: the ability to quickly and accurately carry out mathematical
4. Space: spatial visualizations as well as ability to mentally
transform spatial figures
5. Associative Memory: rote memory
6. Perceptual Speed: quickness in perceiving visual details, anomalies,
7. Reasoning: skill in a variety of inductive, deductive, and arithmetic
So, Thurstone's approach constituted the first multi-factor approach to
Thurstone's tests have largely dropped
out of use because the hope that they would be able to more accurately
predict academic or occupational performance than general intelligence
was not fulfilled.
Nevertheless, the main argument and
findings are important: that intelligence is better described and
measured by considering distinct primary mental abilities, rather than a
single factor g which does not provide specific information about
(see Flanagan, Genshaft & Harrison, 1997; Murphy & Davidshofer, 1998).
9. Raymond Cattell:
Fluid & Crystallised Intelligence (1960’s)
Raymond Cattell (1963) [not to be
confused with James McKeen Cattell, a contemporary of Galton's who was
also significantly involved in early attempts at psychological
measurement] suggested that there are two related but distinct
components of g: fluid and crystallised intelligence.
Fluid: ability to see relationships, as in analogies and letter and
number series = primary reasoning ability
Crystallised: acquired knowledge and skills = factual knowledge
Fluid intelligence decreases with age and crystallised intelligence
increases with age. Thus mathematicians and scientists, who need fluid
intelligence, produce their best work in thier 20s and 30s; whereas
those in the field of history, philosophy and literature produce their
best work in their 40s, 50s and beyond as they have accumulated more
knowledge. Interestingly, poets, who depend more on fluid than
crystallised intelligence, produce their best work earlier than prose
authors: this has been observed in all cultures, languages and
many, many factors! (1960’s - present)
Guilford (1967; 1988) parted company
from the majority of factorial theorists by refusing to acknowledge the
existence of any general factor at all. Instead, he proposed that
intelligence comprises 180 elementary abilities. The 180 elementary
abilities are made up of a combination of three dimensions which he
- operations: what a person does (6-types)
- contents: the material on which operations are performed (5-types)
- products: the form in which the information is stored and processes
Guilford proposed that each combination of a specific operation, a
specific type of content and a specific type of product defines a unique
type of intelligence (6x5x6 = 180). In later versions of his theory he
proposed even more types of intelligence.
Due largely to the practical implications of such a model, Guilford's
theory has not significantly influenced psychological testing of
11 & 12. Vernon & Carroll: Hierarchical Approaches (1960’s - present)
Probably the most widely accepted
factorial description of intelligence is a hierarchical one, e.g.
(1960, 1965, 1971) and
Carroll (1993). Vernon accepted, in a sense, that
both Spearman (single g factor) and Thurstone (multiple primary mental
abilities) were right.
Vernon suggested that intelligence can be described as comprising
abilities at varying levels of generality:
- at the highest level of generality (i.e. top of the hierarchy) is g as
defined by Spearman);
- at the next level are 'major group' factors, such as:
- verbal-educational ability [the kind of ability needed for successful
performance in courses such as English, history, and social studies] and
- practical-mechanical ability [the kind of ability needed successful
performance in courses such as draughtsmanship and car mechanics];
- at the next level are 'minor group' factors, which can obtained by
subdividing the major group factors;
- and at the lowest (the bottom of the hierarchy) are specific factors
again of the kind identified by Spearman.
So, Vernon inserted 2 further levels between Spearman's g and specific
factors relevant to only one test.
Carroll (1993) proposed the three-stratum model of cognitive ability
(similar to Vernon's).
These hierarchical descriptions of intelligence may be viewed as filling
in the gaps between the extreme approaches of Spearman and Thurstone.
Gardner: Multiple intelligences (1980s to the present)
Howard Gardner (1983; 1993) supports Thurstone’s notion that intelligence comes in different packages. The
most widely cited version of Gardner’s concept of intelligence is that
there are seven different types of intelligence. Gardner has played
around with this number and suggested a possible one or two more or even
(in 1999) the possibility of a smaller number of intelligences. He
argues that the seven intelligences are: verbal, mathematical, musical,
spatial, kinaesthetic, interpersonal (social skills) and intrapersonal
(self-understanding) functioning. He argues that these different
intelligences are independent of one another. Critics argue: not
all these things are intelligence: More next week.
Sternberg (1970s to present)
Robert Sternberg (1977, 1985) together
with his colleague Richard Wagner (1993, 1995) argues that there are
Sternberg and Wagner have designed a
test of practical intelligence.