Intelligence Theories and Intelligence Testing

1. Theories of Intelligence

Intelligence is commonly defined as the ability to learn from experience, solve problems, and adapt to the environment. Theories differ on its structure:

• General intelligence (Spearman's g): Charles Spearman found that scores on different mental tasks correlate positively and proposed a single underlying factor, g, that influences performance across all cognitive tasks.
• Fluid vs. crystallized intelligence (Cattell and Horn): Fluid intelligence is the capacity to reason quickly and solve novel problems independent of prior knowledge (peaks in early adulthood); crystallized intelligence is accumulated knowledge, vocabulary, and skills (grows across the lifespan).
• Sternberg's triarchic theory: Intelligence has three aspects — analytical (problem solving and academic skills), creative (generating novel ideas), and practical ("street smarts," adapting to everyday contexts).
• Gardner's multiple intelligences: Intelligence is not one ability but at least eight relatively independent intelligences (e.g., linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, intrapersonal, naturalist). The theory is popular in education but criticized for weak empirical support and unclear boundaries with talent or personality.

2. History and Tools of Intelligence Testing

• Alfred Binet (early 1900s) created the first practical intelligence test to identify schoolchildren needing extra help, measuring mental age. He warned against viewing the score as a fixed, inborn quantity.
• Stanford-Binet: Lewis Terman's American adaptation introduced the intelligence quotient (IQ), originally the ratio of mental age to chronological age times 100.
• Wechsler scales: The WAIS (Wechsler Adult Intelligence Scale) and WISC (Wechsler Intelligence Scale for Children) are the most widely used modern tests. They yield an overall IQ plus index scores (e.g., verbal comprehension, working memory, perceptual reasoning, processing speed).
• IQ: In modern tests, IQ is a standardized score with a mean of 100 and a standard deviation of about 15, so roughly two-thirds of people score between 85 and 115.

3. Test Quality and Fairness

A good test must be standardized — administered and scored the same way for everyone, with norms (scores of a large, representative comparison group) that let an individual's score be interpreted relative to peers. Quality is judged by reliability (consistency of scores across time and forms) and validity (whether the test measures what it claims), including predictive validity (how well it forecasts relevant outcomes such as grades or job performance). Cultural fairness and test bias are persistent concerns: if items or norms favor one cultural or socioeconomic group, the test may measure opportunity and familiarity as much as ability. A test can be statistically reliable yet still produce group differences driven by bias in item content, norming samples, or testing conditions — a reminder that "valid" is always relative to a purpose and population.

How it works

1. Test authors define the construct (intelligence) and write many candidate items.
2. The test is standardized on a large, representative sample to build norms.
3. An individual's raw score is converted to an IQ relative to same-age peers.
4. Psychometricians evaluate reliability and validity, including predictive validity.
5. Users interpret scores cautiously, attending to cultural fairness and possible test bias.
6. Conclusions about nature/nurture respect the limits of heritability and avoid treating any score as fixed potential.

Common confusions

Memory aids

Remember "Gardner Grows Many Skills; Spearman Sees One g" for the theories, and "S-R-V" for test quality — Standardized and Normed, then judged by Reliability and Validity. For the WAIS/WISC pair, recall "A for Adult, C for Child."

Quick review

Topic Recap

Intelligence is theorized as a single general factor (g), as fluid vs. crystallized abilities, as Sternberg's analytical/creative/practical triad, or as Gardner's multiple intelligences. Modern tests (Stanford-Binet, WAIS, WISC) standardize scores into an IQ with mean 100 and SD 15, and their worth depends on standardization, norms, reliability, and validity — including predictive validity. Cultural fairness and test bias are unresolved concerns, and the nature/nurture question must respect the limits of heritability.

Knowledge Check

1. Which theory proposes a single general factor, g, underlying all cognitive abilities?
2. A test gives the same score when taken twice. Which quality does this reflect?
3. The ability to solve novel problems independent of prior knowledge is called what?
4. Which test is used to measure intelligence in adults?
5. Why must heritability estimates be interpreted with caution?

Answers and Rationales

1. Spearman's general intelligence — factor analysis showed positive correlations across tasks, suggesting one underlying factor.
2. Reliability — consistency across administrations is the definition of reliability.
3. Fluid intelligence — novel, knowledge-independent reasoning; crystallized intelligence is accumulated knowledge.
4. WAIS (Wechsler Adult Intelligence Scale) — the WISC is the children's version.
5. Because heritability is a population statistic about variance, not a statement about any individual's fixed potential — it also changes with the environment and cannot explain between-group differences.