Revision Library
Choose a subject
Menu
GCSE Statistics · Topic guide

Skewness

Skewness is a measure of how asymmetrical a data distribution is, that is, whether values are bunched towards one end with a longer tail stretching towards the other. A distribution is positively skewed when mean is greater than median and negatively skewed when mean is less than median; it is symmetrical when neither tail dominates. GCSE Statistics Higher tier judges skewness using summary statistics and diagrams.

HigherSummarising DataEdexcelAQACCEA

Before you start

Make sure you're comfortable with these topics first:

Method

  1. Compare the mean and median: if mean is greater than median the data is positively skewed; if mean is less than median it is negatively skewed; if they are close, the data is roughly symmetrical.
  2. Alternatively, compare the two halves of the interquartile range: find Q2 - Q1 (spread of the lower half) and Q3 - Q2 (spread of the upper half).
  3. If Q3 - Q2 is greater than Q2 - Q1, the data is positively skewed; if Q3 - Q2 is less than Q2 - Q1, it is negatively skewed; if they are equal, the distribution is symmetrical.
  4. On a box plot, check whether the box and whisker sections are longer on the right of the median (positive skew) or the left (negative skew).
  5. On a histogram, check which side has the longer tail: a tail stretching to the right means positive skew, a tail stretching to the left means negative skew.
  6. State the type of skew clearly (positive, negative or symmetrical), referencing the evidence used.

Common mistakes

  • MistakeNaming the skew by where most of the data sits, so that most values low with a long tail of high values is called negative skew.

    InsteadName the skew by the direction of the long tail. A tail towards the high values is positive skew, where the mean is usually greater than the median; a tail towards the low values is negative skew.

  • MistakeWorking out Q1 - Q2 or Q2 - Q3 and comparing negative numbers, or comparing Q3 - Q1 with the median.

    InsteadFind the two half-widths as positive distances, Q2 - Q1 and Q3 - Q2, and compare them. The longer half shows which side of the median the middle 50% is more spread out.

  • MistakeFinding the quartiles from raw data without putting it in order first.

    InsteadOrder the data, find the median, then find Q1 and Q3 as the medians of the lower and upper halves (leaving out the middle value when there is an odd number of values). Only then compare Q2 - Q1 with Q3 - Q2.

Worked examples

Example 1

A set of exam marks has minimum 20, lower quartile 45, median 58, upper quartile 64, maximum 70, and mean 54. Describe the skewness of the distribution.

  1. Compare the mean and median: mean = 54, median = 58, so mean is less than median.
  2. Compare the two halves of the interquartile range: Q2 - Q1 = 58 - 45 = 13 and Q3 - Q2 = 64 - 58 = 6.
  3. Since Q3 - Q2 (6) is less than Q2 - Q1 (13), the upper half of the data is less spread out than the lower half, which indicates negative skew.
  4. Both methods agree: the mean-median comparison and the quartile comparison both point to negative skew.
  5. Final answer: the distribution of exam marks is negatively skewed.

Example 2

The times, in minutes, that 11 pupils took to finish a puzzle were: 3, 5, 7, 7, 7, 8, 10, 14, 17, 22, 30. (a) Find the median and quartiles, leaving out the median when finding them, and use Q2 - Q1 and Q3 - Q2 to describe the skew. (b) Find the mean and say whether it agrees. (c) Suppose the slowest pupil had taken 90 minutes instead of 30. State how each method's verdict changes.

  1. The data is in order and n = 11. The median is the 6th value, 8. The lower half is 3, 5, 7, 7, 7, so Q1 = 7. The upper half is 10, 14, 17, 22, 30, so Q3 = 17.
  2. Q2 - Q1 = 8 - 7 = 1 and Q3 - Q2 = 17 - 8 = 9. Since 9 is greater than 1, the upper half is more spread out, which means positive skew.
  3. The total is 130, so the mean is 130 / 11 = 11.82 (2 d.p.). This is greater than the median of 8, so the mean-median method also gives positive skew and the two methods agree.
  4. If the slowest time is 90, the order of the values is unchanged, so the median (8), Q1 (7) and Q3 (17) are the same and the quartile comparison is unchanged.
  5. The total becomes 130 - 30 + 90 = 190, so the mean becomes 190 / 11 = 17.27 (2 d.p.). The gap between mean and median widens but the verdict does not change.
  6. Final answer: the data is positively skewed by both methods (Q2 - Q1 = 1 and Q3 - Q2 = 9; mean 11.82 above median 8). A time of 90 minutes changes the mean to 17.27 but leaves the quartiles alone, and the skew stays positive.

Example 3

The marks of 12 pupils in a spelling test, out of 15, were: 11, 4, 13, 2, 15, 9, 6, 13, 4, 12, 5, 14. (a) Put the marks in order and find the median and quartiles. (b) Use Q2 - Q1 and Q3 - Q2 to describe the skew. (c) Find the mean and say whether it agrees.

  1. Put the marks in order: 2, 4, 4, 5, 6, 9, 11, 12, 13, 13, 14, 15. There are 12 values.
  2. With 12 values the median is halfway between the 6th and 7th values: (9 + 11) / 2 = 10. The lower half is the first six values, 2, 4, 4, 5, 6, 9, so Q1 = (4 + 5) / 2 = 4.5. The upper half is the last six, 11, 12, 13, 13, 14, 15, so Q3 = (13 + 13) / 2 = 13.
  3. Q2 - Q1 = 10 - 4.5 = 5.5 and Q3 - Q2 = 13 - 10 = 3. Since 3 is less than 5.5, the lower half is more spread out, which means negative skew.
  4. The total is 108, so the mean is 108 / 12 = 9. This is less than the median of 10, so the mean-median method also gives negative skew and the two methods agree.
  5. Final answer: median 10, Q1 = 4.5, Q3 = 13; Q2 - Q1 = 5.5 is greater than Q3 - Q2 = 3 and the mean 9 is below the median 10, so the marks are negatively skewed by both methods.

Practice questions

Q1[1 mark]A distribution has mean 40 and median 40. What type of skew does this suggest?Show answer

Answer: Symmetrical (no skew), since the mean equals the median

Got it right?
Your answerWork it out on paper, then show the answer and mark yourself.
Q2[1 mark]A distribution has mean 62 and median 55. What type of skew does this suggest?Show answer

Answer: Positive skew, since mean (62) is greater than median (55)

Got it right?
Your answerWork it out on paper, then show the answer and mark yourself.
Q3[2 marks]A data set has Q1 = 10, median = 18, Q3 = 22. Compare the two halves of the interquartile range and state the type of skew.Show answer

Answer: Negative skew (Q2 - Q1 = 8 is greater than Q3 - Q2 = 4, so the lower half is more spread out)

Got it right?
Q4[1 mark]A box plot has a short whisker on the left of the box and a long whisker on the right, with the median line close to the left edge of the box. What type of skew is this?Show answer

Answer: Positive skew (the longer tail/whisker is on the right)

Got it right?
Q5[2 marks]A data set has minimum 5, Q1 = 12, median 15, Q3 = 30, maximum 34, and mean 18.5. Using both the mean-median comparison and the quartile comparison, describe the skewness.Show answer

Answer: Positively skewed (mean 18.5 is greater than median 15; Q3 - Q2 = 15 is greater than Q2 - Q1 = 3, so both methods agree)

Got it right?
Q6[2 marks]Class X has mean 68 and median 70. Class Y has mean 70 and median 68. Compare the skewness of the two classes' test score distributions.Show answer

Answer: Class X is negatively skewed (mean less than median); Class Y is positively skewed (mean greater than median), so they are skewed in opposite directions

Got it right?
Your answerWork it out on paper, then show the answer and mark yourself.

Exam-style questions

Written in the style of a GCSE Statistics exam paper, with a full mark scheme.

Exam tips

  • Give the evidence with the verdict: quote the two values or distances you compared, and name the skew from the direction of the longer tail.
  • Describe the skew in the words of the context as well as by name, for example 'most times are short, with a few very long ones'.
Q1[2 marks]A distribution has mean 33 and median 41. State, with a reason, the type of skew shown.Show mark scheme

Tick each line you got.

Nothing ticked yet - 2 available

Got it right?
Q2[4 marks]A data set of delivery times has lower quartile 8 minutes, median 13 minutes, upper quartile 15 minutes. (a) Calculate Q2 - Q1 and Q3 - Q2. (b) State, with a reason, the type of skew shown by these quartiles.Show mark scheme

Tick each line you got.

Nothing ticked yet - 4 available

Got it right?
Q3[5 marks]The times, in minutes, taken by a group of runners to finish a race have mean 52, median 47, lower quartile 40 and upper quartile 50. (a) State whether the mean-median comparison suggests positive or negative skew, giving a reason. (b) Calculate Q2 - Q1 and Q3 - Q2 and state whether this comparison agrees with part (a). (c) Give one reason the two methods might disagree for a real data set.Show mark scheme

Tick each line you got.

Nothing ticked yet - 5 available

Got it right?

See real GCSE Statistics past-paper questions, with official mark schemes 

This topic in real past papers

We have tagged some boards' past papers question by question (not every board or year yet). This topic comes up in:

Free printable worksheet

Want more practice on paper? Download the Skewness worksheet pack - 25 pages of exam-style questions with a full mark scheme. Free to download, no card or account. Print it for personal and classroom use.

This topic is chapter 3 of GCSE Statistics Higher Essentials 2, the whole course as one free printable PDF. Its worksheet is also chapter 3 of GCSE Statistics Higher Topic Workbook 5 and chapter 13 of GCSE Statistics Higher 10-Minute Tests 2, free printable practice books.

Other versions of this worksheet:

Keep going

Not quite what you needed?

Tell us what is missing on Skewness, or which topic to write up next.

Build a full practice pack.