Lesson 2 of 8 · 18 min

IQ scores and the normal curve

Read an IQ on the normal curve with mean 100 and standard deviation 15: calculate z-scores and percentile ranks, and use the 68–95–99.7 rule to find the percentage of people between scores.

What you will be able to do

  • Describe the distribution of IQ scores as normal, with mean 100 and standard deviation 15.
  • Calculate the z-score of an IQ, and the IQ for a given z-score.
  • Use the 68–95–99.7 rule to find the percentage of people above, below or between IQ scores, and percentile ranks.
  • Compare scores on scales with different means and standard deviations using z-scores.

Give a well-made IQ test to thousands of people of the same age and plot their scores. Most land near the middle, fewer further out, and very few at the extremes, in a symmetrical bell shape: a normal distribution. Test makers choose the scale so that the mean is 100 and the standard deviation 15 in every age group.

IQ scores are normally distributed with mean 100 and standard deviation 15. The numbers are the percentages of people in each band: the 68–95–99.7 rule split either side of the mean.

A normal distribution is completely described by its mean and standard deviation, so every IQ can be read as a position on the curve. The key number is how many standard deviations a score lies from the mean: its z-score.

z-score on the IQ scale
z=x−xˉs=IQ−10015z = \frac{x - \bar{x}}{s} = \frac{\text{IQ} - 100}{15}

A positive z-score is above the mean and a negative one below it. Rearranged, the same formula turns a z-score back into an IQ: IQ=100+15z\text{IQ} = 100 + 15z.

Worked example

Convert between IQ scores and z-scores

  • Simple familiar

Full-scale IQs on the WAIS-IV have mean 100 and standard deviation 15.

(a) Mia scores 127. Calculate her z-score and say what it means. (b) Leo's score has a z-score of −1.4-1.4. Calculate his IQ.

5 steps

Try it yourself first, then check the working one step at a time.

Method

  1. z=(IQ−100)÷15z = (\text{IQ} - 100) \div 15.

  2. IQ=100+15z\text{IQ} = 100 + 15z.

  3. Keep the sign: below the mean is negative.

Watch out for

  • Dividing by 100 instead of 15.

  • Losing the negative sign, which gives 121 for Leo.

What is the z-score of an IQ of 85?

Key idea

The 68–95–99.7 rule

In a normal distribution about 68% of scores lie within 1 standard deviation of the mean, about 95% within 2, and about 99.7% within 3.

On the IQ scale: about 68% of people score between 85 and 115, about 95% between 70 and 130, and about 99.7% between 55 and 145.

The curve is symmetrical, so each percentage splits evenly about 100. Half of 68% is 34% between 100 and 115. The band from 115 to 130 is half of 95%−68%95\% - 68\%, which is 13.5%; from 130 to 145 is half of 99.7%−95%99.7\% - 95\%, which is 2.35%; and beyond 145 is half of 100%−99.7%100\% - 99.7\%, which is 0.15%. Add the bands you need.

Interactive

The IQ scale

Drag the score to read its z-score and the percentage of people below and above it. Drag the range handles to read the percentage between two scores.

Worked example

Use the 68–95–99.7 rule for a range of IQ scores

  • Complex familiar

A large school has 1200 students. Assume their IQs follow the WISC-V norms (mean 100, standard deviation 15). Use the 68–95–99.7 rule.

(a) What percentage of the students would you expect to score between 70 and 115? (b) About how many students is that?

4 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Convert each end of the range to a z-score.

  2. Split the range at the mean (100).

  3. Each side: half of 68%, 95% or 99.7%.

  4. Add; multiply by the group size for a number of people.

Watch out for

  • Adding 95% and 68% whole instead of half of each.

  • Giving 978.0 or a decimal number of students: round to whole people.

Use the 68–95–99.7 rule. What percentage of people score above 130 on the WAIS-IV?

in %

Percentile ranks. Test reports often give a percentile rank: the percentage of the norm group who scored below the person. An IQ of 100 has a percentile rank of 50. An IQ of 115 is above the 50% below the mean and the 34% between 100 and 115, so its percentile rank is about 50+34=8450 + 34 = 84.

Use the 68–95–99.7 rule to find the percentile rank of an IQ of 130.

Common mistake

A percentile rank is not a percentage correct

The 97th percentile means scoring higher than about 97% of the norm group, not answering 97% of the items correctly. Someone at the 50th percentile is exactly average.

Comparing scores on different scales. A Wechsler subtest gives a scaled score with mean 10 and standard deviation 3, while an index score has mean 100 and standard deviation 15. Raw numbers on the two cannot be compared, but z-scores can: each says how far the person sits from their age group's mean, in standard deviations. Older editions of the Stanford–Binet used a standard deviation of 16, so the same IQ meant a slightly different position on the curve.

Worked example

Compare a subtest score with an index score

  • Complex familiar

Josh, aged 11, was assessed on the WISC-V. His Working Memory Index is 88 (mean 100, SD 15). On the matrix reasoning subtest, part of a different index (Fluid Reasoning), he has a scaled score of 13 (mean 10, SD 3).

Use z-scores to decide which result is relatively stronger, and explain why the raw numbers cannot be compared directly.

4 steps

Try it yourself first, then check the working one step at a time.

Method

  1. Find each score's z-score using its own scale's mean and SD.

  2. The larger z-score is the relatively stronger result.

Watch out for

  • Using mean 100 and SD 15 for the subtest score.

  • Concluding 88 is better because it is a bigger number.

The extremes. About 2% of people score 130 or above (2.5% by the rule), the level often used as one guide to intellectual giftedness. About 2% score 70 or below. A score around that level is one part of the evidence for an intellectual disability, but it is never enough on its own: the DSM-5-TR (2022) also requires difficulties in everyday adaptive functioning (communicating, looking after oneself, getting on with others) that began during childhood or adolescence, and it allows about 5 points either way for measurement error. A person is never diagnosed from a number alone.

Use the 68–95–99.7 rule. What percentage of people score between 85 and 130?

in %

Exam tip

Show the bands

Sketch the curve, mark 55, 70, 85, 100, 115, 130 and 145, and write 34, 13.5, 2.35 and 0.15 in the bands. Then add the bands you need and show the sum. If a question asks for a z-score, show the formula with numbers in it.

Summary

  • IQ scores are normally distributed with mean 100 and standard deviation 15.

  • z=(IQ−100)÷15z = (\text{IQ} - 100) \div 15, and IQ=100+15z\text{IQ} = 100 + 15z. Negative z-scores are below the mean.

  • About 68% of people score 85–115, 95% score 70–130 and 99.7% score 55–145. The bands either side of 100 are 34%, 13.5%, 2.35% and 0.15%.

  • A percentile rank is the percentage of the norm group scoring below a score: 115 is about the 84th percentile, 130 about the 97.5th.

  • z-scores compare results on scales with different means and standard deviations, such as Wechsler subtests (mean 10, SD 3) and index scores (mean 100, SD 15).

Finished the lesson?

Try the 4 quick checks above and the lesson completes itself, or mark it complete now.

Lesson 2 of 80% read

Save your progress with a free account