AP Syllabus focus: 'If a distribution is symmetric, the mean and median are close. If skewed right or left, the mean is pulled toward the longer tail.'
When a distribution is not balanced, its measures of center no longer line up in the same way. Understanding how mean and median respond to shape helps you interpret data more accurately.
Why Shape Affects Center
When describing a quantitative distribution, the mean and the median both identify center, but they do so differently. The mean uses every value in the data set, so each observation contributes to where it falls. The median is the middle ordered value, so it depends on position rather than the exact size of every observation. Because of this difference, the two measures do not always tell the same story when a distribution has a long tail on one side.
That contrast becomes especially important when a graph is balanced on both sides or stretched farther in one direction.
Symmetric Distributions
A distribution is called symmetric when its two sides are approximately balanced around the center.
Symmetric distribution: A distribution whose left and right sides are approximately mirror images around the center.
In a symmetric distribution, values above the center are balanced by values below the center. That balance keeps the mean and median close together. In a perfectly symmetric distribution, they may be exactly equal, but real data are usually only approximately symmetric. For that reason, AP Statistics responses should usually say the mean and median are close or approximately equal, not necessarily identical.
Many real distributions are not balanced.
They are skewed instead, meaning one tail stretches farther than the other.
Skewed distribution: A distribution with a longer tail on one side than the other; skewed right has a longer right tail, and skewed left has a longer left tail.
The direction of skew is determined by the longer tail, not by the side containing most of the data. A distribution is skewed right if the tail extends toward larger values and skewed left if the tail extends toward smaller values.
Why the Mean Is Pulled
The mean is often described as being pulled toward the longer tail because extreme values directly affect an average.
A few very large observations raise the mean, and a few very small observations lower it. The median does not ignore those values, but it changes only when the middle position in the ordered data changes. As a result, the mean usually moves farther toward the tail than the median does.
This is the central idea for interpreting skewed distributions: the mean follows unusual values more strongly than the median.
Right-Skewed Distributions
In a right-skewed distribution, most observations are concentrated at lower or moderate values, while a smaller number of large values create a long right tail. Those larger observations increase the mean more than they increase the median. Therefore, the mean is usually greater than the median.
On a graph, the mean tends to lie farther to the right, closer to the stretched-out side of the distribution. This is why a right-skewed distribution is often described as having the mean pulled to the right.
Do not assume the difference must be large. A slight right skew may produce only a small separation between the two measures, but the overall pattern still places the mean above the median.
Left-Skewed Distributions
In a left-skewed distribution, the longer tail extends toward smaller values. A relatively small number of low observations pull the mean downward, while the median remains closer to the middle of the bulk of the data. Therefore, the mean is usually less than the median.
The stronger the left skew, the more noticeable this separation is likely to be, although the exact gap depends on the particular data set. The key idea is direction: the mean is pulled toward the long left tail.
Writing Statistical Interpretations
On AP Statistics questions, strong responses connect the graph’s shape to the relationship between the two measures of center. You should describe both the direction of skew and the effect on the mean.
If a distribution is roughly symmetric, say the mean and median are close.
If a distribution is skewed right, say the mean is greater than the median because high values in the right tail pull the mean upward.
If a distribution is skewed left, say the mean is less than the median because low values in the left tail pull the mean downward.
Use careful language such as approximately, tends to, or usually when describing real data.
Tie the statement to the displayed distribution rather than giving the rule with no reference to shape.
In other words, do not just memorize a direction rule. Link the visual shape of the distribution to the numerical relationship between mean and median.
Common Misunderstandings
Several common mistakes weaken statistical descriptions.
The side with the longer tail determines the skew, even if most bars or points are on the opposite side.
Mean and median being close does not prove perfect symmetry; it only supports that description.
The median can still change in a skewed distribution; it is simply less affected by extreme values than the mean.
A single unusual value can affect the mean without making the entire distribution strongly skewed.
When justifying a claim, describe the distribution first and then state what that implies about mean and median.
FAQ
Rounding can hide small differences. For example, two statistics that differ by only a few hundredths may both round to the same tenth or whole number.
This means a calculator output might show matching rounded values even though the unrounded mean and median are not exactly the same. In mild skew, that happens fairly often.
In small samples, one unusual observation can shift the mean quite a bit, so the gap between mean and median may jump around from sample to sample.
In larger samples, the overall shape is usually more stable. The mean-median relationship still reflects skew, but it is less likely to be driven by a single point.
Removing a very large value usually lowers the mean noticeably because the mean directly uses that value in the average.
The median may not change at all, especially if the removed value was far from the middle of the ordered data. That is why checking unusual points can strongly affect the mean-median comparison.
Yes. If values cannot go above a fixed maximum, a long right tail may be cut off, even if the underlying process would have produced one.
That can make the distribution look less right-skewed than expected or even create left skew near the upper limit. In bounded data, always consider whether the scale itself is shaping the distribution.
With two clear peaks, the main story of the data may be the split into two groups rather than simple skew. Mean and median each compress the distribution into one number, so neither captures that separation well.
In that situation, the difference between mean and median may still exist, but it does not describe the most important structural feature of the graph.
Practice Questions
A distribution of waiting times is skewed right. State whether the mean is greater than or less than the median, and explain why.
1 mark: States that the mean is greater than the median.
1 mark: Explains that the larger values in the right tail pull the mean upward or to the right.
A student says, “If the mean of a distribution is less than the median, then the distribution is skewed right.”
Evaluate the student’s claim. In your response, describe the relationship between mean and median for:
a symmetric distribution,
a right-skewed distribution,
a left-skewed distribution,
and explain why the relationship changes when a distribution has a long tail.
1 mark: States that the student’s claim is incorrect.
1 mark: States that in a symmetric distribution, the mean and median are close or approximately equal.
1 mark: States that in a right-skewed distribution, the mean is greater than the median.
1 mark: States that in a left-skewed distribution, the mean is less than the median.
1 mark: Explains that the mean uses all values and is pulled toward the longer tail, while the median depends on ordered position and is less affected.
