13.2 Measurement and Assessment of Intelligence

History and Development of Intelligence Testing

Evolution of intelligence testing

Intelligence testing began as a practical tool for education and has since grown into a broad field of cognitive assessment.

Binet-Simon Scale (1905) was the first practical intelligence test. Alfred Binet and Théodore Simon designed it to identify French schoolchildren who needed extra academic support. It introduced the concept of mental age, the idea that a child's intellectual performance could be compared to what's typical for a given age.

Stanford-Binet Intelligence Scales (1916) came next. Lewis Terman at Stanford adapted the Binet-Simon work for American populations and introduced the Intelligence Quotient (IQ) as a single numerical score. This made it easy to compare individuals, which drove widespread adoption.

Wechsler Scales (1939) were developed by David Wechsler, who argued that the Stanford-Binet relied too heavily on verbal ability. His tests separated scores into verbal and performance (nonverbal) scales, giving a more rounded picture of cognitive function. The Wechsler scales also replaced the mental-age ratio with a deviation IQ, comparing a person's score to others in the same age group.

Modern assessments have continued to diversify. Tests like the Kaufman Brief Intelligence Test, Raven's Progressive Matrices, and the Cognitive Abilities Test each target different aspects of cognition or serve different practical needs (quick screening, nonverbal assessment, group testing in schools).

Key Concepts and Types of Intelligence Tests

Fundamentals of test design

Four properties determine whether an intelligence test is trustworthy and useful:

• Standardization means every person takes the test under the same conditions, with the same instructions, time limits, and scoring rules. Without it, you can't meaningfully compare one person's score to another's.
• Reliability is consistency. A reliable test produces similar results when the same person takes it again (test-retest reliability) and when different parts of the test agree with each other (internal consistency). If a test gives you wildly different scores each time, it isn't measuring much of anything.
• Validity is whether the test actually measures what it claims to measure. There are several types:
  • Content validity: Do the test items genuinely sample the domain of intelligence?
  • Construct validity: Do the results align with established theories of intelligence?
  • Predictive validity: Do scores predict real-world outcomes, like academic or job performance?
• Norms are the scores from a large, representative reference group. Your raw score means little on its own; norms let you see where you fall compared to others of the same age.

Types of intelligence assessments

• Individual tests are administered one-on-one by a trained examiner. Examples include the Wechsler Adult Intelligence Scale (WAIS) and the Stanford-Binet. These allow the examiner to observe behavior, ask follow-up questions, and get a detailed profile, but they're time-consuming and expensive.
• Group tests are given to many people at once, making them efficient for schools and large organizations. The Otis-Lennon School Ability Test and the Cognitive Abilities Test (CogAT) are common examples. The tradeoff is less individual observation and flexibility.
• Verbal tests assess language-based skills like vocabulary, reading comprehension, and verbal reasoning. They work well for people fluent in the test's language but can disadvantage non-native speakers.
• Nonverbal tests minimize language demands and instead use visual-spatial reasoning, pattern recognition, and abstract problem-solving. Raven's Progressive Matrices, for instance, presents a series of visual patterns with a missing piece, and you pick the one that completes the pattern. These are often used when language or cultural background could bias results.

Interpretation and Implications of Intelligence Testing

Interpretation of test scores

IQ scores are the most familiar metric. Modern IQ is a deviation score, not a ratio. The score is set so the population mean is 100 and the standard deviation is 15. This means about 68% of people score between 85 and 115, and about 95% score between 70 and 130.

Percentile ranks tell you the percentage of the norm group that scored at or below a given score. A percentile rank of 75 means you scored as well as or better than 75% of the comparison group. The 50th percentile is average.

Age equivalents compare a person's performance to the typical performance of people at various ages. For example, if a 7-year-old performs like the average 9-year-old, their age equivalent is 9. These have limited utility because cognitive development isn't uniform across skills or ages, so they can be misleading.

Ethics in intelligence testing

Intelligence tests carry real consequences for people's lives, which makes ethical considerations essential.

• Educational settings: Test scores can determine placement in gifted programs or special education. The risk is that biased test content or administration can mislabel students, and those labels can become self-fulfilling prophecies where teachers and students adjust expectations to match the score rather than the student's actual potential.
• Employment contexts: Using intelligence tests in hiring raises legal questions. Tests must be demonstrably relevant to job performance, and employers must watch for adverse impact, where a test disproportionately screens out members of a particular group without a valid job-related reason.
• Cultural considerations: Many tests were developed and normed on specific populations. Items that assume particular cultural knowledge or language fluency can systematically disadvantage people from different backgrounds. Creating truly culture-fair tests remains an ongoing challenge.
• Privacy and confidentiality: Test results are sensitive personal information. Ethical practice requires secure storage, limited access, and informed consent about how results will be used.
• Limitations: No intelligence test captures the full range of human cognitive ability. Factors like motivation, test anxiety, creativity, and practical problem-solving are largely missed. Treating a single score as a complete picture of someone's intellect is a misuse of the tool.