From: veritasium
The concept of intelligence testing dates back centuries [00:01:22]. A significant breakthrough occurred in 1904 [00:01:24].
Charles Spearman and the G-Factor
In 1904, English psychologist Charles Spearman began studying students’ grades across different subjects [00:01:29]. He investigated how performance in one subject, such as English, related to performance in another, like Math [00:01:34].
Spearman found a clear positive correlation: students who excelled in Math also tended to perform well in English, with a correlation coefficient of 0.64 [00:03:17]. This pattern was consistent across all subjects, including Classics and French [00:03:35]. To explain this, Spearman proposed that each person possesses a level of general intelligence, which he termed the “g-factor” [00:03:48]. This g-factor was conceptualized as the ability to quickly learn new material, recognize patterns, and think critically across various subjects [00:03:56]. He published his findings in a paper titled “General Intelligence” Objectively Determined and Measured [00:04:15].
Since the correlations were not perfect, Spearman also introduced “s-factors,” or subject-specific factors [00:04:25]. A student’s performance in a subject like Math, for example, would be a combination of their general intelligence and their specific aptitude for Math [00:04:32]. While he believed specific factors could be improved through training, he viewed general intelligence as fixed [00:04:45]. This led him to seek a reliable method for measuring general intelligence [00:04:51].
Alfred Binet and the Birth of the IQ Test
Around the same period, in France, Alfred Binet was tasked with identifying students who needed additional academic support [00:04:57]. Collaborating with Theodore Simon, he developed the Binet-Simon test [00:05:05]. This test included 30 tasks, such as identifying missing items in drawings, defining abstract terms, repeating sentences, and even evaluating which face was prettiest [00:05:10].
Performance on the Binet-Simon test was benchmarked against the abilities of average students at different ages to determine a “mental age” [00:05:25]. For instance, if a student performed like an average eight-year-old, their mental age would be eight [00:05:32]. The mental age was then divided by the student’s actual age and multiplied by 100 to yield an “intelligence quotient,” or IQ [00:05:39]. The Binet-Simon test was thus the world’s first IQ test [00:05:50].
Expansion and Standardization in the U.S.
The Binet-Simon test was translated into English by Henry Goddard and brought to the United States [00:05:54]. At Stanford, Lewis Terman further standardized it using a large American sample, leading to the creation of the Stanford-Binet test [00:06:00]. For decades, the Stanford-Binet test was the most widely used IQ test in the U.S. [00:06:09].
Many other IQ tests were subsequently developed, all with the shared aim of measuring the g-factor [00:06:16]. These tests assess a range of mental abilities, including memory, verbal, spatial, and numerical skills [00:06:23]. The idea was that by averaging scores across multiple sections, subject-specific variations would cancel out, providing a good approximation of general intelligence [00:06:36]. Psychologists designed tests with 7 to 10 distinct sections to minimize subject-specific distortions [00:06:48].
To standardize scoring across different tests, raw scores are normalized, typically with a mean of 100 and a standard deviation of 15 [00:07:03]. This standardization means that about 68% of people have an IQ between 85 and 115, while only about 2% score above 130 or below 70 [00:07:28].
Types of Questions on Modern IQ Tests
Modern IQ tests commonly feature several types of questions:
- Vocabulary: Tests often present a word and require selecting a synonym or antonym from multiple-choice options [00:07:56].
- Number Patterns: These sections require identifying the next number in a sequence by recognizing underlying patterns, which are often simple arithmetic progressions or multiplication series [00:08:28].
- Raven’s Progressive Matrices: A well-known type of question involving a three-by-three grid of symbols where test-takers must select the ninth symbol that completes the pattern [00:09:33]. Common logical rules for these puzzles include translational motion, rotational motion, missing symbols (where each symbol appears once per row/column), and addition (where lines may cancel out if they overlap) [00:09:46].
Most modern IQ tests are administered under time pressure, with typically 10 to 30 seconds per question [00:10:31].
Dark History and Misinterpretations
Despite their widespread use, IQ tests have a controversial history, particularly in the United States. When Henry Goddard introduced Binet’s test to America, its application and interpretation diverged significantly from Binet’s original intent [00:16:50]. Binet believed intelligence could be enhanced through education and designed his tests to identify students needing more help [00:16:57]. However, in the U.S., modified versions of the test were used on adults to rank them by intelligence [00:17:07]. Researchers like Spearman maintained that the g-factor was unchangeable and largely inherited [00:17:16].
These assertions led Henry Goddard to integrate IQ tests into the American eugenics movement, which sought to prevent individuals with “undesirable traits” from procreating [00:18:16]. Laws were enacted in many states allowing forced sterilization of individuals who scored below a certain IQ threshold [00:18:33]. The constitutionality of these laws was upheld by the Supreme Court in 1927 [00:18:42]. Justice Oliver Wendell Holmes famously stated, “Three generations of imbeciles are enough” [00:19:08]. Over 60,000 people were forcibly sterilized as a result, and these laws even served as a model for Nazi Germany [00:19:14].
This awful history contributes to why many people disregard IQ tests today [00:19:41]. Early researchers made several errors, including the belief that IQ is entirely determined by genetics [00:19:53]. Modern research indicates that IQ is influenced by both genetics and environment, with estimates ranging from 40% to 70% heritability [00:20:02]. Additionally, education can improve IQ, demonstrating that it is not completely fixed throughout a lifetime [00:20:29].
IQ tests have also been misused to promote ideas of racial differences in intelligence, despite evidence suggesting that cultural changes and environmental factors significantly affect average scores [00:21:06]. The “Flynn Effect” demonstrates that average IQ scores have increased by approximately 30 points over the last century due to factors like improved nutrition, health, education, and shifts in work requiring more abstract thinking [00:22:05]. This suggests that IQ tests do not objectively measure inherent intelligence across all populations and cultures [00:23:50].