My grandfather was
clever. With no formal schooling to speak of, he could build anything,
from a dollhouse to a real house, from scratch, without plans. He also
could fix anything -- kitchen appliances, cars, children's toys, radios,
televisions, you name it. He even published a book of his poems when he
was in his 70s. He was not clever, however, at taking I.Q. tests, which
he confronted in grade school, in the military, and when he looked for
a job when he was in his early 20s. He hated taking the tests; he was made
anxious by the clock ticking as he worked, and he found it confusing and
unnatural to think in terms of abstractions, be they mathematical, pictorial,
or verbal.
Because of his performance on tests, my grandfather did not consider himself
very intelligent. Neither did the teachers, military recruiters, and job-placement
personnel who used the test scores: They reduced my grandfather's intelligence
to a simple, relatively low number on a page and labeled him "slow." The
I.Q. tests that my grandfather took in the 1930s -- versions of which are
still in use today -- were created to determine which children failing
in school were doing so because of low intelligence, and which were failing
for other reasons. Through questions about the meaning of words or paragraphs,
mathematical problems, visual patterns, and so forth, these tests measured
intelligence in terms of the number of problems a person could solve, compared
with the average for other people of the same age.
Throughout our society, we still use I.Q. tests, and their close surrogates
such as the SAT, in the belief that they provide a meaningful measure of
a person's innate intelligence and capacity for success in intellectual
tasks. We all know the considerable weight these tests are given throughout
education, as well as in hiring and promotion decisions in the workplace.
But scholars still have not explained how, if I.Q. tests tell us the most
important things we need to know about a person's intelligence, we can
account for my grandfather and the many others like him, who are competent
and successful in so many domains in the real world.
This is the issue that my colleagues and I have studied in our attempt
to democratize the concept of intelligence, by including in it more and
different types of abilities and talents. While we have been conducting
our research, other scholars working in the same area have demonstrated
that I.Q. tests' reputation as an ultimate seal of approval was premature.
For example, consider the work of James Flynn, a political scientist at
the University of Otago, in New Zealand. He proved that I.Q. scores have
risen sharply over the past 60 or more years in all 20 nations for which
data exist. In fact, a person born in 1877 whose score put him or her in
what was then the 90th percentile on a widely used reasoning test would,
with exactly the same number of correct answers, rank in only the 5th percentile
of people born in 1967. (Flynn proved this by examining the raw numbers
of correct answers on the same tests used over time. Most researchers rely
on "normed scores," which are adjusted to keep the average score on a test
constant from year to year, and which thus cannot accurately be compared
over time.)
Are we really that much smarter than our grandparents? How could I.Q. scores
change so much, so quickly?
We learned two things from Flynn's work: First, a high I.Q. score does
not necessarily mean intelligence, nor does a low score mean stupidity.
Second, whatever the test measures is highly mutable. Flynn is fond of
saying that, if we take I.Q. scores seriously as meaningful predictors
of intelligence, our grandparents would have been unable to understand
the rules of baseball. Given the rapidity of the changes Flynn reported,
genetics could not be responsible, and so researchers have focused on aspects
of culture, as well as on health and nutrition, in attempting to explain
why people today are markedly outscoring their ancestors.
One possible cultural factor is that people are increasingly familiar with
the material on certain types of I.Q. tests. My grandfather's generation
rarely encountered anything in their everyday lives even remotely resembling
the items on such tests. Today, however, mazes, puzzles, and other games
that are thinly disguised versions of items from actual I.Q. tests appear
on cereal boxes and on placemats at fast-food restaurants. People play
with toys such as Rubik's Cube. Some computer screen-saver programs are
strikingly similar to other kinds of intelligence tests: The complex patterns
dancing
around the screen closely resemble the Raven matrices, the most popular
test of reasoning ability. Is it any wonder that today's kids outperform
my grandfather's generation?
But the more important question is: Does this greater exposure to material
similar to that on the tests make today's children and adults smarter in
any meaningful way than earlier generations of test takers? I think not.
The intellectual accomplishments of people in past eras are awe-inspiring,
and the challenges and hardships that they had to overcome were extraordinary.
Looking back on these accomplishments should make us cautious in interpreting
the significance of I.Q. scores as predictors of likely success in the
real world.
Perhaps the reason that so many individuals with low or moderate I.Q.'s,
such as my grandfather, are so successful in their daily lives can be found
in recent research that has broadened the concept of intelligence. Researchers
today are demonstrating empirically the importance of many abilities that
are not measured on I.Q. tests. Consider studies that my colleagues and
I have conducted to assess practical and creative thinking in business,
the military, and elementary and middle schools.
We wanted to know why some business managers with M.B.A.'s from prestigious
graduate schools alienate their subordinates virtually overnight, why some
military leaders lose the respect of their soldiers and subordinate officers
by adhering to formal doctrines even in situations where they are not adequate,
and why some bright children hand in boring compositions after the deadline
and then react with surprise when they receive low grades. We found that
all of these people lack practical intelligence -- an ability essential
to success that differs from the more "academic" intelligence measured
by I.Q. tests, and which is largely independent of it.
We learned that practical intelligence consists of three types of abilities
-- managing oneself, managing others, and managing the organization or
environment in which one works, such as a school, corporation, or hospital.
Each ability is important in a unique way, and each contributes to real-world
success. People may be strong in one type of practical intelligence and
weak in another, although, generally, being savvy about managing organizations
builds on the abilities to manage oneself and others. Importantly, traditional
measures of I.Q. tell us little about who has and does not have the three
types of practical intelligence.
Where are scientists headed in our search to understand intelligence? Increasingly,
we think in terms of types and facets of intelligence that lead to success
in specific contexts: social intelligence, emotional intelligence, creative
intelligence, and practical intelligence. We look at people's ability to
manage their lives by motivating and organizing themselves to perform effectively.
We consider people's ability to get along with their employees, peers,
supervisors, and teachers. Often, it is those types of intelligence, as
much as I.Q. scores, that determine success or failure in education and
in the workplace, especially among people with a similar range of I.Q.
scores.
Historically, a person's intelligence was reduced to a single number. Today,
that number still holds sway in many admissions offices, but the realization
is growing that we need to characterize and measure more of the abilities
that are important to adult success. We owe the next generation a broader
and more relevant battery of tests, designed to measure the many varied
abilities that contribute to success in the real world. Better tests will
lead to the admission of applicants with a wider variety of skills, thus
diversifying further the pool of talent available to our society.
As we look ahead to the demographic changes under way and recognize the
need to distribute educational and employment opportunities fairly and
broadly, it becomes even more essential for us to assess people's capabilities
accurately. We need a conception of intelligence that encompasses my grandfather's
talents. The most successful leaders in business, the professions, and
other enterprises know how to define workable goals and motivate themselves
to accomplish them; they know how to "read" and motivate other people;
and they know how to distinguish solutions that work in the real world
from ones that work only in books -- all abilities that current I.Q. tests
do not measure.
This is not to say that success on an I.Q. test does not provide meaningful
information; it is just that other types of success matter, too. It should
not escape us that the technological developments on which our society
depends may require types of intelligence -- practical and creative, for
example -- that are different from those emphasized in our standardized
tests. The science of understanding intelligence thus may progress farther
and faster by recognizing the wisdom of our grandparents.
----------------------------
Wendy M. Williams
is an associate professor of human development at Cornell University and
the co-author, with Stephen J. Ceci, of Escaping the Advice Trap, just
published by Andrews McMeel.