Intelligence Testing

Learning Objectives

By the end of this page, you will be able to:

• Trace the origin of modern intelligence testing from Binet to the Stanford-Binet and Wechsler scales.
• Compare Spearman's g-factor, Gardner's Multiple Intelligences, and Sternberg's Triarchic Theory.
• Explain how raw scores become standardized IQ scores using a normal distribution.
• Identify the major standardized intelligence tests and what each is designed to measure.
• Evaluate the major criticisms of intelligence testing, including cultural bias and narrow scope.
• Apply intelligence testing concepts to educational, clinical, and occupational scenarios.

Quick Answer

Intelligence testing is the attempt to measure cognitive ability — reasoning, problem-solving, memory, processing speed — using standardized tasks, then expressing performance as a score (an IQ) relative to same-age peers. It matters because IQ scores are among psychology's most-used and most-debated tools: they inform special education placement, disability diagnosis, and cognitive research, while also drawing sustained criticism for cultural bias and for equating a narrow set of tasks with the broad, messy concept of "intelligence." Understanding both what these tests measure and what they leave out is essential to using them responsibly.

History of Intelligence Testing

Definition: The historical development of standardized instruments designed to quantify individual differences in cognitive ability.

Explanation: In 1905, French psychologist Alfred Binet, working with Théodore Simon, built the first practical intelligence test — not out of theoretical curiosity, but to solve a real problem: French schools needed a way to identify children who needed extra academic support. Binet's test compared a child's performance to what was typical for their age, producing a "mental age." Lewis Terman later adapted and expanded Binet's test at Stanford University, creating the Stanford-Binet Intelligence Scale and introducing the ratio IQ formula: (mental age / chronological age) × 100.

Example: A child with a mental age of 8 and a chronological age of 10 would have a ratio IQ of (8/10) × 100 = 80.

Real-world example: Modern school psychologists still trace their instruments back to Binet's basic insight — comparing an individual's performance against an age-matched norm group — even though the ratio-IQ formula itself has been replaced by more statistically sound methods (below).

Why it matters: Binet's original purpose — identifying children who need support — is easy to lose sight of once intelligence testing becomes associated with ranking or labeling people. Its origin as a practical, supportive tool is worth remembering when weighing its appropriate use today.

Common misunderstanding: Students think Binet set out to rank all of humanity by intelligence. He explicitly warned against treating his scale as a fixed measure of innate, unchangeable capacity — a caution the field didn't always heed in the decades that followed.

Theories of Intelligence

Spearman's g-Factor Theory

Definition: The theory that a single general intelligence factor, "g," underlies performance across all cognitive tasks.

Explanation: Charles Spearman noticed that people who scored well on one type of cognitive test tended to score well on others too — a pattern he explained by proposing one general factor (g) contributing to all tasks, alongside task-specific factors (s).

Example: Someone strong in vocabulary tasks also tending to do well on spatial puzzles is exactly the correlational pattern that led Spearman to propose g.

Real-world example: Most modern IQ tests, including the WAIS, still report a single composite score (Full Scale IQ) partly because of the empirical support for a general factor underlying performance across subtests.

Why it matters: The g-factor is one of the most replicated findings in psychology and forms the statistical backbone of most standardized intelligence tests today.

Common misunderstanding: Students think "g" means intelligence is one single, indivisible thing. Even Spearman's own model includes task-specific factors (s) — g explains shared variance across tasks, not the whole story.

Gardner's Theory of Multiple Intelligences

Definition: The theory that intelligence is not one general ability but several relatively independent types, including linguistic, logical-mathematical, spatial, musical, bodily-kinesthetic, interpersonal, and intrapersonal intelligence.

Explanation: Howard Gardner argued that traditional IQ tests overvalue verbal and logical-mathematical skills while ignoring abilities — like musical talent or bodily coordination — that are just as cognitively demanding and culturally valuable.

Example: A gifted musician who struggles with standardized math tests would, in Gardner's framework, be recognized as highly intelligent in a domain traditional IQ tests don't measure at all.

Real-world example: Many progressive schools design curricula around Gardner's framework, offering varied paths (art, movement, music) for students to demonstrate understanding, not just written tests.

Why it matters: Gardner's theory broadened public and educational conversations about what "smart" means, even though it remains controversial as empirical science.

Common misunderstanding: Students treat Gardner's theory as equally empirically supported as g-factor research. In fact, critics point out that Gardner's "intelligences" often lack the statistical intercorrelation evidence that supports g, and some (like "naturalistic" intelligence) function more as talents or interests than measurable cognitive abilities.

Sternberg's Triarchic Theory

Definition: A theory proposing three distinct components of intelligence: analytical, creative, and practical.

Explanation: Robert Sternberg argued traditional tests measure mainly analytical intelligence (the kind rewarded in classrooms), while creative intelligence (generating novel solutions) and practical intelligence ("street smarts," adapting to real-world environments) are just as important but rarely tested.

Example: A person who scores modestly on a standardized test but consistently solves ambiguous, real-world workplace problems well is displaying strong practical intelligence.

Real-world example: Sternberg's own research found that practical intelligence measures predicted job performance about as well as, or better than, traditional IQ tests in some occupational settings.

Why it matters: This theory highlights a real limitation of standardized testing — that success in life draws on abilities beyond what a timed, analytical test can capture.

Common misunderstanding: Students assume "practical intelligence" is just a synonym for "common sense" that anyone naturally has. Sternberg treats it as a genuine, measurable cognitive skill that some people develop more than others through experience.

Standardized Tests

Definition: Instruments administered and scored in a fixed, consistent manner so scores can be meaningfully compared across individuals.

Explanation: The two major families of individually administered intelligence tests are the Stanford-Binet and the Wechsler scales. The Wechsler Adult Intelligence Scale (WAIS) and its child version, the Wechsler Intelligence Scale for Children (WISC), both produce a Full Scale IQ along with index scores for domains like verbal comprehension, perceptual reasoning, working memory, and processing speed — giving a profile, not just a single number.

Example: A WAIS profile might show high verbal comprehension but low processing speed, a pattern that could suggest something clinically meaningful (like an attention difficulty) rather than uniformly "low" or "high" ability.

Real-world example: Raven's Progressive Matrices, a non-verbal test of pattern reasoning, is often used specifically because it minimizes reliance on language and cultural knowledge — useful when testing across linguistic or cultural groups.

Why it matters: Modern IQ scores are calculated as deviation IQs — standard scores based on how far a person's performance falls from the mean of their age group on a normal distribution (mean 100, standard deviation 15) — rather than the old mental-age ratio, because deviation scoring works sensibly for adults (whose "mental age" doesn't meaningfully keep climbing after a certain point).

Common misunderstanding: Students still describe IQ using the old ratio formula (mental age / chronological age × 100). Modern tests use deviation IQ scoring instead — a score of 115 means a person scored one standard deviation above the mean for their age group, not that their "mental age" exceeds their real age by some fixed ratio.

Interpreting Intelligence Test Results

Interpretation blends two layers: quantitative (the standardized IQ score and index scores) and qualitative (patterns of relative strength and weakness across domains). A Full Scale IQ of 100 tells you a person performed at the population average, but the index-score profile often matters more clinically — a large gap between verbal and processing-speed scores can point toward a specific learning difficulty even when the overall IQ looks unremarkable.

Why it matters: Treating the Full Scale IQ as the only meaningful output ignores exactly the kind of pattern that often triggers a referral for further evaluation (e.g., a learning disability assessment).

Limitations and Criticisms

• Cultural bias — Tests built around one cultural group's vocabulary, references, and problem-solving conventions can systematically disadvantage test-takers from other backgrounds, even when their underlying ability is equivalent.
• Narrow definition of intelligence — Critics argue IQ tests miss creativity, emotional intelligence, and practical problem-solving — exactly the gap Gardner's and Sternberg's theories try to address.
• Overemphasis on verbal ability — Early tests leaned heavily on language skills, which could penalize non-native speakers or those with language-based learning differences independent of general cognitive ability.
• Stability over time — IQ scores are reasonably stable in adulthood but can shift meaningfully across childhood and in response to test conditions, motivation, and practice effects.

Why it matters: These criticisms don't mean IQ tests are useless — they remain strong predictors of academic and occupational outcomes on average — but they do mean scores must be interpreted with humility and always in context, never as a definitive verdict on someone's worth or potential.

Real-World Applications

• Education — Identifying students for gifted programs or for learning-disability services (often via a discrepancy between ability and achievement).
• Employment — Cognitive ability tests are used in some hiring pipelines, generally alongside other selection tools rather than alone.
• Clinical psychology — Diagnosing intellectual disability requires both a low IQ score and deficits in adaptive functioning — a two-part criterion that guards against over-relying on the number alone.
• Research — Studying cognitive aging, the effects of interventions, or group differences in cognition all depend on having a standardized way to measure "ability" in the first place.

Key Terms

Term	Definition
IQ (Intelligence Quotient)	A standardized score representing cognitive ability relative to an age-matched norm group.
Deviation IQ	Modern IQ scoring method: a standard score with mean 100 and standard deviation 15, based on a normal distribution.
g factor (general intelligence)	Spearman's proposed single factor underlying performance across diverse cognitive tasks.
Multiple Intelligences	Gardner's theory that intelligence comprises several relatively independent domains (linguistic, spatial, musical, etc.).
Triarchic Theory	Sternberg's theory of three intelligence components: analytical, creative, and practical.
Full Scale IQ	The single composite score summarizing overall performance across an intelligence test's subtests.
Index score	A sub-score representing performance in a specific cognitive domain (e.g., working memory) on tests like the WAIS.
Cultural bias (in testing)	Systematic disadvantage in test performance caused by content favoring one cultural or linguistic group.

Common Mistakes

Misconception 1: "IQ is calculated as mental age divided by chronological age times 100." Why it's wrong: That ratio-IQ formula was Terman's original method and breaks down in adulthood, since mental age doesn't keep increasing indefinitely. Correct understanding: Modern tests use deviation IQ — a standard score derived from where a person falls on a normal distribution relative to their age group (mean 100, SD 15).

Misconception 2: "IQ tests measure a single, unified thing called intelligence, and that's the end of the debate." Why it's wrong: Spearman's g-factor has strong empirical support for shared variance across tasks, but theories like Gardner's Multiple Intelligences and Sternberg's Triarchic Theory highlight real abilities (creativity, practical problem-solving, musical talent) that standard IQ tests don't capture. Correct understanding: g explains a great deal of the correlation across cognitive tasks, but "intelligence" as a lived, practical concept is broader than what any single test captures.

Misconception 3: "A low IQ score alone is sufficient to diagnose an intellectual disability." Why it's wrong: Clinical diagnosis of intellectual disability requires a low IQ score and significant deficits in adaptive functioning (daily living skills, communication, social skills). Correct understanding: The score is one piece of evidence; a full diagnostic picture always requires assessing functional impact, not the number alone.

Comparison and Connections

Theory	Core Claim	Key Limitation
Spearman's g-factor	One general ability underlies all cognitive tasks	Doesn't fully explain domain-specific talents
Gardner's Multiple Intelligences	Several independent types of intelligence exist	Limited statistical/empirical support; hard to measure objectively
Sternberg's Triarchic Theory	Analytical, creative, and practical intelligence are distinct	Practical/creative components are harder to standardize than analytical ones

Scoring Method	Formula/Basis	Problem It Solved
Ratio IQ (Terman)	(Mental age / chronological age) × 100	Simple but breaks down for adults
Deviation IQ (modern)	Standard score, mean 100, SD 15	Works consistently across all ages

Practice Questions

Recall

1. Who developed the first practical intelligence test, and why was it created? Answer guidance: Alfred Binet (with Théodore Simon) in 1905, to identify French schoolchildren who needed additional academic support.
2. What are the mean and standard deviation used in modern deviation IQ scoring? Answer guidance: Mean = 100, standard deviation = 15.

Understanding

3. Explain why the ratio-IQ formula doesn't work well for adults. Answer guidance: Mental age plateaus in adulthood while chronological age keeps increasing, so the ratio would artificially decrease with age even if ability stayed constant — deviation IQ avoids this by comparing performance to same-age peers instead.
4. Compare Gardner's and Sternberg's critiques of traditional IQ testing. Answer guidance: Gardner argues intelligence comprises multiple independent domains (linguistic, spatial, musical, etc.) that traditional tests ignore; Sternberg argues traditional tests overemphasize analytical ability while neglecting creative and practical intelligence — both agree standard IQ tests are too narrow, but propose different alternative structures.

Application

5. A non-native English speaker scores unexpectedly low on a verbally loaded IQ test. What limitation of intelligence testing does this illustrate, and what alternative might address it? Answer guidance: This illustrates cultural/linguistic bias in verbally loaded tests; a non-verbal instrument like Raven's Progressive Matrices could provide a fairer estimate of fluid reasoning ability.
6. A clinician is evaluating a child for intellectual disability and finds a low WISC Full Scale IQ. What additional information must be gathered before making a diagnosis? Answer guidance: Evidence of deficits in adaptive functioning (daily living skills, communication, social skills) — a low IQ score alone is insufficient for a diagnosis.

Analysis

7. Evaluate the claim: "Because g is one of the most replicated findings in psychology, alternative theories like Multiple Intelligences should be dismissed." Answer guidance: This overstates the case — g explains substantial shared variance across cognitive tasks, but doesn't preclude the existence of specific talents or abilities (musical, interpersonal, practical) that aren't well captured by tasks used to measure g; the theories address different questions and are not mutually exclusive that easily.
8. A company wants to use a WAIS-style test to screen job applicants. Analyze the potential risks of doing so without complementary assessment methods. Answer guidance: Risks include cultural/linguistic bias against some applicant groups, over-reliance on analytical ability while ignoring creative/practical skills relevant to the job (per Sternberg), and treating a single score as decisive rather than one input among several (interviews, work samples, references).

FAQ

Q1: Is IQ fixed for life once measured? No. IQ is reasonably stable in adulthood but can shift during childhood and adolescence, and scores are also affected by test conditions, motivation, practice effects, and, in some cases, targeted interventions.

Q2: Why do IQ tests report multiple index scores instead of just one number? Because a single Full Scale IQ can mask meaningful differences — someone with average overall IQ might have a large gap between verbal and processing-speed scores, a pattern relevant to identifying learning disabilities.

Q3: Are non-verbal tests like Raven's Progressive Matrices "bias-free"? Not entirely — no test is perfectly culture-free — but minimizing language and culturally specific content reduces one major source of bias, making them useful for cross-cultural or multilingual testing contexts.

Q4: Why did Terman's ratio-IQ formula get replaced? Because mental age effectively stops climbing in adulthood while chronological age keeps increasing, making the ratio meaningless for adults; deviation IQ scoring fixed this by comparing scores within age-matched norm groups.

Q5: Does a high IQ guarantee success in life? No. IQ correlates with academic and occupational outcomes on average, but Sternberg's and Gardner's work highlights that creative and practical abilities, motivation, and opportunity all contribute substantially to real-world success independent of IQ.

Quick Revision

• Binet (1905) created the first intelligence test to identify students needing academic support, not to rank people.
• Terman's Stanford-Binet introduced ratio IQ: (mental age / chronological age) × 100.
• Modern tests use deviation IQ: standard score, mean 100, SD 15.
• Spearman's g factor = one general ability underlying performance across tasks.
• Gardner's Multiple Intelligences = several independent domains (linguistic, spatial, musical, etc.).
• Sternberg's Triarchic Theory = analytical, creative, and practical intelligence.
• WAIS (adults) and WISC (children) produce a Full Scale IQ plus index scores across domains.
• Raven's Progressive Matrices reduces verbal/cultural bias via non-verbal pattern reasoning.
• Diagnosing intellectual disability requires low IQ AND adaptive functioning deficits — not the score alone.
• Major criticisms: cultural bias, narrow scope, overemphasis on verbal ability.
• Interpretation should combine quantitative scores with qualitative profile patterns.

Related Topics

Prerequisites: Introduction to Psychological Assessment (assessment process and categories).

Related Topics: Personality Assessment (a parallel standardized-testing domain); Neuropsychological Testing (uses cognitive testing to detect brain-related impairment).

Next Topics: Personality Assessment — the next major category of standardized psychological testing.