• Practice
  • About 15 min
  • 5 practice questions

The normal curve and IQ: z-scores, percentages and conversions

Use the bell-shaped distribution of IQ scores with mean 100 and standard deviation 15 to find z-scores, the share of people within a range, and conversions between two scales.

If something looks wrong, please tell us.

Many intelligence-test questions are calculations, not definitions. They use the fact that scores on a well-made test are spread in a bell shape around the average, the normal curve. If you know the mean and the standard deviation, a few rules let you answer most such questions in one line. This lesson gives the rules and several worked conversions.

The rules

  • The normal curve is symmetric: half the people are above the mean and half below.
  • The standard deviation (SD) measures spread. On Wechsler-type IQ scales the mean is 100 and the SD is 15. Other tests may use a different SD, so read the question.
  • About 68% of people lie within 1 SD of the mean, about 95% within 2 SD and about 99.7% within 3 SD.
  • The z-score of a score X is z = (X − mean)/SD. It says how many SDs the score is above (positive) or below (negative) the mean.
  • To convert a z-score to the IQ scale, use IQ = 100 + 15z.
RegionShare of people
Between the mean and 1 SD above (or below)about 34%
Between 1 SD and 2 SD above (or below)about 13.5%
More than 2 SD above (or below)about 2.5%

z = (X − mean)/SD. Within 1 SD: about 68%. Within 2 SD: about 95%. Beyond 2 SD on one side: about 2.5%.

Worked example 1: a score of 130

On a scale with mean 100 and SD 15, what is the z-score of IQ 130, and what share of people score higher? z = (130 − 100)/15 = 2. The score is 2 SD above the mean, and about 2.5% of people score above that (the other 2.5% of those outside 2 SD score below 70). So roughly 2 or 3 people in every 100 score higher.

Worked example 2: ranges

The middle 68% of people have IQs within 1 SD of 100: from 100 − 15 to 100 + 15, that is 85 to 115. The middle 95% lie from 100 − 30 to 100 + 30, that is 70 to 130. Notice that about 16% (half of the remaining 32%) score below 85 and about 16% score above 115.

Worked example 3: converting from another scale

A test has mean 50 and SD 10. A student scores 65. The z-score is (65 − 50)/10 = 1.5. On the IQ scale (mean 100, SD 15) the same position is 100 + 15 × 1.5 = 122.5. For a raw score of 40 on the same test, z = −1, which is an IQ of 85. Converting through z-scores works for any two scales, because z measures position in standard deviations.

Worked example 4: a group of 1000

Suppose, as an illustration, 1000 children are tested on the IQ scale and the scores are normally distributed. About 68% of 1000 = 680 children score between 85 and 115. About 95% = 950 score between 70 and 130. About 2.5% of 1000 = 25 score above 130, and the same number below 70. These are approximations, and real groups differ a little from the ideal curve.

What the numbers do not say

An IQ score is an estimate with some error, and labels such as 'gifted' or 'below average' use cut-offs that differ between sources. Never treat a single score as a final verdict on a child. Use it with other evidence such as classroom work and observation.

Key takeaways: use z = (X − mean)/SD, and IQ = 100 + 15z on the Wechsler-type scale. Remember 68%, 95% and 99.7% for 1, 2 and 3 SD. The tails beyond 2 SD hold about 2.5% each. Convert between scales through z. Treat scores as estimates.

Check your understanding

  1. 1. On the IQ scale (mean 100, SD 15), between which two scores do about 68% of people lie?

    1. 70 and 130
    2. 90 and 110
    3. 55 and 145
    4. 85 and 115
    Show answer

    Answer: D. Within 1 SD of 100 means 100 ± 15, that is 85 to 115.

  2. 2. What is the z-score of an IQ of 130 on a scale with mean 100 and SD 15?

    1. 1
    2. 2
    3. 1.5
    4. 30
    Show answer

    Answer: B. z = (130 − 100)/15 = 2.

  3. 3. About what percentage of people score above an IQ of 130 (2 SD above the mean)?

    1. About 5%
    2. About 16%
    3. About 2.5%
    4. About 34%
    Show answer

    Answer: C. About 95% lie within 2 SD, so about 5% lie outside, and half of that, about 2.5%, is above 2 SD.

  4. 4. A test has mean 50 and SD 10. A raw score of 65 is equal to which IQ on the scale with mean 100 and SD 15?

    1. 122.5
    2. 117.5
    3. 125.0
    4. 132.5
    Show answer

    Answer: A. z = (65 − 50)/10 = 1.5, so IQ = 100 + 15 × 1.5 = 122.5.

  5. 5. In a normally distributed group of 1000 children, about how many score between 70 and 130?

    1. About 680
    2. About 997
    3. About 950
    4. About 500
    Show answer

    Answer: C. About 95% of scores lie within 2 SD of the mean, and 95% of 1000 is 950.

Privacy · Terms · Refunds · Delivery