IQ Test Scoring Methods

Ratio IQ: The Original Method

When IQ testing was first developed in the early 20th century, the score was calculated using what is now called ratio IQ. This method was introduced by German psychologist William Stern in 1912 and popularized by Lewis Terman in the original Stanford-Binet test.

The formula was straightforward:

IQ = (Mental Age / Chronological Age) x 100

For example, if a 10-year-old child performed at the level typical of a 12-year-old, their IQ would be (12 / 10) x 100 = 120. If that same child performed at the level of an 8-year-old, the IQ would be (8 / 10) x 100 = 80.

This method worked reasonably well for children, since cognitive development follows a roughly predictable trajectory during childhood. However, it had a critical flaw: it broke down for adults. Cognitive development does not continue increasing linearly after a certain age, so the concept of "mental age" becomes meaningless for adults. A 40-year-old who performs at the level of a 50-year-old is not necessarily more intelligent; mental age plateaus in early adulthood.

Because of this limitation, the ratio IQ method was abandoned by all major IQ tests by the mid-20th century, replaced by the deviation IQ approach described below. For more on the historical development of IQ testing, see our guide on the history of IQ testing.

Deviation IQ: The Modern Standard

The deviation IQ method, introduced by David Wechsler in 1939 with the Wechsler-Bellevue Intelligence Scale, is now used by virtually all major IQ tests. Instead of comparing your performance to an age-based standard, it compares your raw score to the statistical distribution of scores from a large, representative norming sample of people your own age.

The key parameters are:

• Mean (average) score – Set to 100 by definition. If you perform exactly at the average level for your age group, your IQ is 100.
• Standard deviation – Set to 15 on the Wechsler scales and most modern tests, or 16 on the Stanford-Binet (historically, though the SB5 now also uses 15). This determines how far from the mean each score falls in statistical terms.

The resulting score tells you where you rank relative to the population. An IQ of 115 means you scored one standard deviation above the mean, placing you at approximately the 84th percentile. An IQ of 130 is two standard deviations above the mean, placing you at the 98th percentile. An IQ of 85 is one standard deviation below the mean, at approximately the 16th percentile.

This method works equally well for children and adults, since everyone is compared to their own age group. It also produces scores that follow a normal (bell-shaped) distribution, which makes statistical analysis and comparison straightforward.

Wechsler Scoring: Index Scores and FSIQ

The Wechsler scales (WAIS-IV for adults, WISC-V for children) are the most widely used IQ tests in the world, and their scoring system is the one most people encounter in clinical settings.

The WAIS-IV produces five main scores:

• Verbal Comprehension Index (VCI) – Measures verbal reasoning, word knowledge, and the ability to express concepts verbally. Subtests include Similarities, Vocabulary, and Information.
• Perceptual Reasoning Index (PRI) – Assesses nonverbal reasoning, spatial processing, and the ability to analyze visual information. Subtests include Block Design, Matrix Reasoning, and Visual Puzzles.
• Working Memory Index (WMI) – Evaluates the ability to hold information in mind and manipulate it. Subtests include Digit Span and Arithmetic.
• Processing Speed Index (PSI) – Measures the speed at which you can process simple visual information and make decisions. Subtests include Symbol Search and Coding.
• Full Scale IQ (FSIQ) – A composite score derived from all four index scores. This is the number most people think of as "the IQ score." It provides a single summary measure of overall cognitive ability.

Each index score and the FSIQ are calculated using the deviation IQ method with a mean of 100 and standard deviation of 15. Individual subtest scores use a different scale: a mean of 10 and standard deviation of 3, called scaled scores.

The WISC-V (for children) uses a similar structure but includes a fifth index: the Visual Spatial Index (VSI), split from the broader Perceptual Reasoning category. It also adds a Fluid Reasoning Index (FRI). Learn more about IQ testing for children.

Stanford-Binet 5 Scoring

The Stanford-Binet Intelligence Scales, Fifth Edition (SB5) uses a five-factor model to assess cognitive abilities. Each factor is measured with both a verbal and a nonverbal subtest, producing ten subtests total:

• Fluid Reasoning – The ability to solve novel problems using logic, independent of prior knowledge. This factor is closely related to what psychometricians call fluid intelligence.
• Knowledge – Accumulated information and the ability to retrieve and apply it, corresponding to crystallized intelligence.
• Quantitative Reasoning – Numerical problem-solving and the understanding of mathematical concepts.
• Visual-Spatial Processing – The ability to perceive, analyze, and think in terms of visual patterns and spatial relationships.
• Working Memory – The capacity to hold and manipulate information in short-term memory.

The SB5 produces a Full Scale IQ (FSIQ) along with Nonverbal IQ (NVIQ) and Verbal IQ (VIQ) composite scores, plus five Factor Index scores. All are reported on the standard scale with a mean of 100 and standard deviation of 15. Individual subtest scaled scores use a mean of 10 and standard deviation of 3, consistent with the Wechsler system.

One notable feature of the SB5 is its wide age range: it can be administered to individuals from age 2 through adulthood, making it particularly useful for assessing very young children and for tracking cognitive development over time.

Composite Scores vs. Subtest Scores

One of the most important distinctions in IQ test scoring is between composite scores and subtest scores. Understanding this distinction is essential for properly interpreting IQ test results.

Subtest scores are the individual scores you receive on each specific task within the test. On the WAIS-IV, for example, you receive separate scores for Block Design, Vocabulary, Digit Span, and so on. These scores are reported as scaled scores (mean of 10, standard deviation of 3) and reflect your performance on each narrow cognitive task.

Index scores are composites of related subtests. They summarize performance across a broader cognitive domain, such as verbal comprehension or processing speed. These are reported on the standard IQ scale (mean of 100, standard deviation of 15).

The Full Scale IQ is the broadest composite, combining all index scores into a single number. While the FSIQ is the most commonly reported and discussed score, clinicians often find the index and subtest scores more informative, because they can reveal specific patterns of strengths and weaknesses that the overall number obscures.

For example, a person with an FSIQ of 100 might have a Verbal Comprehension Index of 120 and a Processing Speed Index of 80. The composite is average, but the pattern reveals significant verbal strengths coupled with processing speed challenges, which could be clinically meaningful. This is why looking beyond the single number matters, as our guide on whether your IQ score is good explains.

How Online Tests Approximate Scoring

Online IQ tests, including our free test at Take-IQTest.com, cannot replicate the full scoring methodology of a clinical instrument. However, well-designed online tests use a simplified version of the deviation IQ approach:

1. Raw score calculation – Your total number of correct answers is tallied, sometimes with weighting for question difficulty.
2. Normative comparison – Your raw score is compared to a distribution of scores from other test takers or a reference dataset. The comparison group may be based on the test's accumulated user data or calibrated against published norms.
3. Conversion to the IQ scale – Your position in the distribution is converted to a score on the standard scale (mean of 100, standard deviation of 15), giving you an estimated IQ that is comparable to the format used by clinical tests.

The main limitations of this approach are that online tests typically cannot provide meaningful subtest or index scores (since they have fewer questions per domain), and the norming sample may not be as rigorously controlled as the samples used for clinical instruments. Nonetheless, this method produces a useful approximation that correlates meaningfully with formal IQ scores for most people.

For more on evaluating the quality of online IQ tests, see our guide on whether IQ quizzes are legitimate, and for a broader look at what different types of tests measure, visit our page on how IQ is calculated.