How IQ Ranges Are Calculated | SD 15 Explained
Related Reading
An IQ score of 115 seems straightforward until you ask: how was that number determined? What does it actually mean? Behind every intelligence test score lies sophisticated statistical methodology designed to place your cognitive ability within a reliable framework. Understanding how IQ ranges are calculated transforms a mysterious number into meaningful information about your cognitive performance relative to others.
The Standardization System: The Foundation of IQ Ranges
Modern IQ testing relies on standardization—a process that gives test scores consistent meaning across different populations and time periods. Standardization begins with large-scale testing of representative populations. Thousands of people take the test, and statisticians analyze the results to determine how performance distributes.
IQ tests are deliberately calibrated so that the average score is always 100, with a standard deviation of 15 points. This means approximately 68% of the population scores between 85 and 115 (within one standard deviation), and approximately 95% scores between 70 and 130 (within two standard deviations). This bell curve distribution is not accidental—it's mathematically constructed.
How Raw Scores Become IQ Scores
The conversion from raw performance to IQ score involves several steps. First, a test-taker answers questions and receives a raw score—the number of correct answers. This raw score is then converted to an age-adjusted score, because cognitive ability changes across the lifespan. A raw score of 40 questions correct might represent different abilities for a 25-year-old versus a 65-year-old.
The age-adjusted score is then converted to an IQ equivalent using a statistical technique called "norm-referencing." The test developers have established what percentile rank different raw scores represent. If your age-adjusted score places you at the 84th percentile, you receive an IQ score of approximately 115 (one standard deviation above the mean of 100).
Understanding IQ Ranges and What They Represent
- Below 70: Significantly below average (fewer than 3% of population)
- 70-84: Below average cognitive ability (14% of population)
- 85-115: Average cognitive ability (68% of population)
- 116-130: Above average to superior (14% of population)
- Above 130: Exceptionally high cognitive ability (fewer than 3% of population)
These ranges describe relative performance, not absolute intelligence. Someone at 70 is not "70% intelligent"—rather, they perform cognitive tasks at a level lower than 97% of the population. Someone at 130 performs at a level higher than 97% of the population. The ranges represent your position within the distribution of human cognitive ability.
The Role of Percentiles in IQ Interpretation
Percentiles provide more intuitive meaning than raw IQ scores. The 50th percentile represents average performance—half the population scores higher, half scores lower. The 84th percentile means approximately 84% of people score at or below your level. This percentile framework makes IQ scores interpretable without deep statistical knowledge.
Research from the American Psychological Association shows that percentiles are more accurately understood than IQ scores alone. When people hear "you scored at the 84th percentile," they grasp this means above-average performance. When they hear "IQ of 115," the meaning is less immediately clear.
Why the Standard Deviation Matters
The standard deviation of 15 points is critical to how ranges are calculated. Some IQ tests use a standard deviation of 16 (older Stanford-Binet tests). This single difference changes the entire IQ scale. An IQ of 116 on a test with SD=15 is not identical to 116 on a test with SD=16. Understanding which standard deviation a test uses is essential for accurate comparison.
Confidence Intervals: The Uncertainty Built Into IQ Scores
Every IQ score includes measurement error. A score of 115 doesn't mean your true ability is exactly 115—it means your true ability likely falls within a confidence interval, typically 110-120. Modern intelligence test manuals specify these confidence intervals (usually at 90% or 95% confidence levels) to acknowledge measurement imprecision.
This means comparing IQ scores requires caution. An IQ of 110 and 115 don't represent meaningfully different cognitive abilities—the difference falls within measurement error. Reliable differences typically require gaps of at least 5-10 IQ points depending on the test.
Flynn Effect: Why IQ Ranges Have Shifted Over Time
IQ tests must be periodically renormalized because average performance has increased over decades. This phenomenon, discovered by psychologist James Flynn, shows that raw test performance has risen approximately 3 IQ points per decade in developed nations. Tests from the 1950s must be updated to maintain the average at 100.
This requires periodic renorming where new populations take the test and statisticians recalibrate the scale. The Cognitive Index test uses current normative data to ensure that today's average score of 100 reflects contemporary cognitive performance, not performance from outdated reference groups.
Practical Meaning of IQ Ranges
Beyond statistical definitions, what do IQ ranges predict about real-world performance? Research shows moderate to strong correlations between IQ ranges and academic achievement, job performance in complex roles, and learning speed. However, IQ is far from the only predictor of success. Motivation, personality, opportunity, and specific skill development matter tremendously.
Someone at the 84th percentile (IQ ~115) has genuine cognitive advantages in problem-solving and learning speed. But someone at the 50th percentile (IQ ~100) possesses perfectly adequate cognitive ability for most life pursuits. The ranges identify performance differences, not worth differences or potential.
Worked Examples: Turning Scores Into Percentiles
| IQ score | Percentile (SD 15) | Interpretation |
|---|---|---|
| 85 | 16th | Low average; below the midpoint but still within the broad normal range |
| 100 | 50th | Average; right at the norm group mean |
| 115 | 84th | One standard deviation above average; strong above-average reasoning |
| 130 | 98th | Exceptionally high; near the top few percent of the distribution |
These worked examples are useful because they show how IQ scores are not linear labels. The jump from 100 to 115 is not the same as the jump from 115 to 130. Each step reflects a position within a distribution, so the practical meaning comes from percentile rank and confidence interval, not the number alone.
Real-World Example: A Job Candidate Interpreting a Score
A candidate finishes a timed reasoning test and gets an estimated IQ of 118. The number is encouraging, but the real question is what it means in context. On a norm-referenced scale, that result is comfortably above average and may suggest strong analytical potential. But if the candidate’s score comes from a short practice test, the score should be treated as directional rather than definitive. A better interpretation combines the estimate with subskill data, the confidence interval, and repeated performance across multiple sessions.
This same logic applies to students, job seekers, and anyone comparing test results across different publishers. If one test uses a broader spread or a different norm group, the same raw performance can map to a different percentile. That is why context matters more than the label attached to a single score.
Common Misconceptions About IQ Ranges
- An IQ score is not a percentage and does not mean you are 'that percent intelligent.'
- A five-point difference is often too small to interpret without confidence intervals.
- SD 15 and SD 16 scales are not interchangeable without conversion.
- Percentile rank is usually easier to interpret than the raw IQ score itself.
- A single score should never be treated as a complete description of a person’s ability.
FAQ
FAQ
What does an IQ of 115 mean?
On a standard SD 15 scale, 115 is about one standard deviation above the mean and is usually around the 84th percentile.
Why do some tests use SD 16 instead of SD 15?
Different test publishers use different scoring conventions. SD 16 is common in some older or specific test systems, while SD 15 is the modern standard for many assessments.
Is percentile rank better than IQ score?
Percentile rank is often easier for non-specialists to understand, but both can be useful when they are explained correctly and tied to the right norm group.
How much can IQ scores change between tests?
Small differences can happen because of measurement error, timing, and the specific norming method. Differences of a few points are often not meaningful without context.
Can practice improve an IQ-style score?
Practice can improve familiarity, pacing, and error reduction, which can raise scores somewhat even if it does not change the underlying construct being measured.
Conclusion
IQ ranges are calculated through rigorous statistical standardization that gives test scores consistent, comparable meaning. By establishing an average of 100 with a standard deviation of 15, and using percentile conversion, intelligence testing creates an interpretable scale. Understanding this methodology reveals that IQ scores are scientifically grounded, though imperfect measures of cognitive ability. Your IQ range describes your cognitive performance relative to others—valuable information, but only one dimension of your complete cognitive profile.
For more cognitive science research, visit CognitiveIndex.org
See also: Understanding how working memory and general intelligence differ in cognitive ability assessment, and exploring whether logic puzzle practice can increase your measured intelligence.