Why IQ, SAT, and GRE don't measure overall intelligence (GDP)

01 октября 2022 г. в 21:23

By Stephen Ciesi, Cornell University. The book "Introduction to psychology". Authors - R. L. Atkinson, R. C. Atkinson, E. E. Smith, D. J. BEM, S. Nolen-Hoeksema. Under the General editorship of V. P. Zinchenko. 15th international edition, Saint Petersburg, Prime-euroznak, 2007. Article from Chapter 12. Individual differences

There is a funny anecdote about a man waiting for a bus, and when a woman passes by, pushing a baby carriage in front of her, she looks into the stroller.

Man: "Oh, what a beautiful baby you have!"

Woman: "Don't judge by her face, you should look at her photos!»

We all know people who look more interesting in photos than in real life. For such people, the results of intelligence tests (photos) are more important than the skills that these tests are thought to predict (the child).

Thousands of "validity" studies show that General intelligence tests predict a wide range of different behaviors, although not perfectly, but better than any other method known to us. First-year students ' grades are predicted by IQ scores slightly better than the grades or characteristics obtained by students in high school. Students ' grades in their first year of graduate school are also better predicted by IQ scores than by grades and characteristics obtained at University. But the accuracy of predictions based on IQ (or SAT or GRE) is limited, and many candidates ' scores will differ from what is expected. Test makers argue that even limited predictability can help school enrollment officials make better decisions than without using tests (Hunt, 1995).

For many years, I headed the admissions Committee. One of my saddest discoveries was that tests like GRE are given great importance. I admit that there are other reasons for choosing from candidates with similar grades and GRE test results, but candidates with very low test results will not be accepted to most universities unless the institution needs to fill places for financial reasons.

Everyone who works on admissions committees can tell stories about applicants with very high test scores who turn out to be losers, unable to think creatively and do their job effectively. Or about the rare applicants with low test scores who somehow entered an educational institution and refuted the forecast, becoming one of the best students. I'm not saying that General intelligence tests like GRE, IQ, and SAT are useless. But they are not perfect, and applicants who do well with them are only one of the types of people who can do well in school. Other types of applicants (for example, those with well-developed practical skills, creative abilities, and interpersonal skills) are systematically rejected.

But my biggest complaint about General intelligence tests is that they seem to me to be no more a means of measuring General intelligence than achievement tests in math or history. As with all achievement tests, SAT, intelligence, and GRE scores are subject to external influences, including tutoring, the quality of school teaching, the level of education of parents, and their earnings. It will not be a surprise to find that the number of school math classes allows you to predict the result obtained in the mathematical questions included in the SAT and GRE, even after measuring the initial mathematical abilities. This means that while points earned for math questions are added to the score that is used to predict College or graduate school grades, they may not reflect an innate ability to do math any more than a test score in French indicates a person's ability to master that language if they haven't studied French in school. Thus, I often view "General intelligence" tests as tests of achievement and believe that they reflect the amount of knowledge a person has accumulated. They tell us little about actual abilities in various circumstances, especially in situations that require thinking differently than when answering test questions. Such tests also tell us little about important talents that cannot be measured, such as creativity, practical intelligence, the ability to organize, motivate, and manage oneself and others — skills that seem to be somewhat independent of so-called intelligence tests and that have been shown to predict important educational outcomes.

Recently, a large number of studies have been conducted on the predictive validity of SAT and GRE, and it has been found that while the results of these tests can predict grades obtained in the first year of study, their predictive ability decreases over time. One recent study even reported that GRE does not allow predicting second, third, and fourth — year grades or the creativity and quality of a doctoral dissertation after the first year of graduate school (Sternberg & Williams, 1997), although it is possible that this is a "dormant effect" and intelligence indicators predict future career achievements and earnings as a result. These tests, which are actually achievement tests, have many valid applications, but it is unwise to call them "General intelligence" tests until it is shown that they can provide insight into a much wider range of abilities and achievements than is currently available. And until their predictive ability increases, we should not confuse the child and his photos.

The results of the tests SAT and GRE — accurate indicators of intelligence
How can you know that two types of measurement point to the same thing? You calculate a statistical metric called correlation (which varies from 0 to 1.0), and the higher the correlation, the more similar these types of measurements are. Tests of aptitude, achievement, intelligence, and cognitive ability correlate so highly with each other that many experts believe that these tests are all actually the same.


  • Интеллект
  • ВВП. Глава 12
  • Психологические теории

Комментарии (0):