Questions: Intelligence Test Construction and Score Interpretation

Question 1

A psychologist administers a WAIS test normed in 1980 to a patient in 2010 and reports a score of 110. What problem does the Flynn effect reveal with this score interpretation?

• A. The test items have become too easy through cultural exposure, so the patient likely solved them by recall rather than reasoning
• B. The 1980 norms are outdated — the patient's raw score would correspond to a lower score on current norms, meaning 110 likely overstates ability relative to today's population
• C. The patient should be penalized for items that reflect cultural knowledge not available in 1980
• D. The score is valid because IQ is an absolute measure of cognitive ability that does not change with historical period

Answer: (B)

The Flynn effect documents a secular rise of approximately 3 IQ points per decade in raw scores across populations. If norms were established in 1980, the raw score needed to achieve 100 on that test is lower than what the average person today would score. Using old norms inflates IQ scores — the patient may look above average compared to the 1980 cohort but be near average relative to a current normative sample. This is why major intelligence batteries are periodically renormed. Option D is the core misconception the Flynn effect refutes.

Question 2

A test developer reports that a 10-year-old child has a 'mental age' of 12. A psychologist argues this ratio IQ approach is inferior to deviation IQ. What is the most important reason?

• A. Mental age is a subjective concept that cannot be operationalized, while deviation IQ is purely mathematical
• B. Ratio IQs (MA/CA × 100) become uninterpretable for adults because cognitive development slows while chronological age keeps increasing, distorting the score
• C. Mental age conflates intelligence with academic achievement, while deviation IQ separates the two constructs
• D. Mental age scores cannot be compared across different intelligence tests, while deviation IQ is universal

Answer: (B)

The ratio IQ loses meaning in adulthood because raw score growth decelerates sharply after adolescence while chronological age keeps increasing — producing artificially declining scores in older adults. A 40-year-old scoring at the 'mental age' of a 35-year-old would get an IQ below 90, misrepresenting their actual cognitive standing. Deviation IQ solves this by asking: where does this person fall in the distribution for their age group? A score of 115 means one SD above the mean for one's age, regardless of whether the test-taker is 20 or 65.

Question 3

A person's raw score on an intelligence test is meaningless for interpretation without a normative reference sample from a representative population.

• T. True
• F. False

Answer: True

Unlike a temperature reading (which has a physical referent) or a percentage correct (which has a defined maximum), raw scores on intelligence tests have no intrinsic interpretation. Getting 68 items correct on the WAIS says nothing about cognitive ability unless you know the distribution of scores in a representative sample. The norming process converts raw scores into deviation IQ scores (mean 100, SD 15) by establishing where a given raw score falls in an age-matched normative distribution — this is why test manuals specify the norming sample's size, demographics, and date.

Question 4

Because intelligence tests are grounded in a hierarchical factor model with g at the apex, an individual's performance on one composite index reliably predicts their performance on most other indices.

• T. True
• F. False

Answer: False

The existence of g does not mean cognitive abilities are uniform within a person. The CHC model explicitly recognizes broad abilities at the second stratum — Verbal Comprehension, Perceptual Reasoning, Working Memory, Processing Speed — that are correlated but distinct. An individual can score in the superior range on Verbal Comprehension and average on Processing Speed. Clinicians use the profile across indices diagnostically; significant discrepancies between indices are often more informative than the overall composite. The misconception conflates g (a statistical factor explaining shared variance) with the absence of meaningful ability differences within individuals.

Question 5

What is the difference between a deviation IQ and a ratio IQ, and why did psychometricians shift to the deviation method?

Model answer: A ratio IQ divides mental age (the age level at which a person scores) by chronological age and multiplies by 100. A deviation IQ compares a person's raw score to the distribution of scores in an age-matched normative sample, expressing the result as a standard score with mean 100 and standard deviation 15. The shift occurred because ratio IQs become uninterpretable in adulthood: raw score growth decelerates after adolescence while chronological age keeps increasing, producing artificially declining scores. Deviation IQ asks 'where does this person fall relative to their age peers?' — a question that remains meaningful across the lifespan.

The deviation approach also produces scores with known statistical properties (fixed mean and SD) that are directly comparable across age groups and, if scale parameters match, across tests. The ratio IQ lacks these properties — its distribution varies with age and its SD is not constant, making comparisons unreliable. Modern intelligence batteries universally use deviation IQ for these reasons.