Skew in AP Statistics

In AP Statistics, skew describes a distribution that is not symmetric because one tail stretches out farther than the other. It is skewed right if the long tail points toward larger values and skewed left if it points toward smaller values, and the mean usually gets pulled toward the tail.

What is Skew?

Skew is the word you use when a distribution is lopsided. Most of the data piles up on one side, and a long, thin tail trails off toward the other side. You name the skew after the tail, not the pile. If the tail stretches toward bigger numbers, the distribution is skewed right (positive skew). If the tail stretches toward smaller numbers, it's skewed left (negative skew). Think of the tail as pointing at the direction of the skew.

The CED ties skew directly to how the mean and median behave (EK 1.8.B.1). In a roughly symmetric distribution, the mean and median sit close together. In a right-skewed distribution, the mean is usually larger than the median. In a left-skewed one, it's usually smaller. Here's why. The median only cares about the middle value, so a few extreme numbers in the tail barely move it. The mean adds up every value, so those tail values drag it toward them. That makes the mean like a rope that the tail tugs on.

Why Skew matters in AP® Statistics

Skew lives in Unit 1, Topic 1.8 (Graphical Representations of Summary Statistics). It supports learning objective AP Stats 1.8.B, which asks you to describe distributions based on the relationship between the mean and median, and AP Stats 1.8.A, which asks you to construct graphs like boxplots from the five-number summary. Shape is one of the four things you always describe for a distribution (shape, center, variability, and unusual features), and skew is the most common shape you'll talk about. It also decides which summary statistics make sense. With strong skew, the median and IQR describe the data more honestly than the mean and standard deviation. That same idea comes back later in the course when you check whether it's safe to run inference procedures.

How Skew connects across the course

Box Plot (Unit 1)

A boxplot is a picture of the five-number summary, so you can read skew right off it. If one whisker is much longer, or the median sits closer to one end of the box, the data is stretched toward the longer side. A right-skewed boxplot has a long upper whisker, and a left-skewed boxplot has a long lower whisker.

Outlier (Unit 1)

Outliers often sit out in the tail of a skewed distribution, and they make the mean-versus-median gap even bigger. This is why the median and IQR are called resistant. They barely move when a few extreme values show up, while the mean and standard deviation get yanked toward the outliers.

Descriptive Statistics (Unit 1)

Skew is the reason you can't pick summary statistics blindly. For a roughly symmetric distribution, the mean and standard deviation work well. For a strongly skewed one, the median and IQR give a better sense of a typical value and the spread, so the shape you see should guide the numbers you report.

Normal Distribution (Units 1, 4)

A normal distribution is perfectly symmetric, so skew is basically a measure of how far data strays from that bell shape. That matters in Unit 4. When you run inference for means with a small sample, you check that the data shows no strong skew or outliers before you trust a t-procedure.

Is Skew on the AP® Statistics exam?

Skew shows up constantly, usually in three ways.

1. Reading shape from a graph. You'll get a histogram, dotplot, or boxplot and need to name the shape. Released FRQs that used the term include 2019 FRQ Q1 (a histogram of room sizes in a residence hall) and 2018 FRQ Q5 (histograms of teaching year at two high schools). On these, a full-credit description names the shape in context, alongside center, variability, and any unusual features.

2. Mean vs. median reasoning. MCQs love asking what happens to summary statistics when a distribution is skewed or has outliers. A typical stem gives you right-skewed housing prices with several high-value outliers and asks about the effect on the summary statistics. The answer is that the mean is pulled above the median. Another stem asks which measures of center and spread are least affected by high outliers, and the answer is the median and IQR.

3. Interpreting boxplots and five-number summaries. You might see a five-number summary (like min = 15, Q1 = 22, median = 30, Q3 = 35, max = 52) and need to judge the shape by comparing whisker lengths and where the median sits in the box. A boxplot with no upper whisker means the maximum equals Q3, so the top quarter of the data is stacked at one value.

Comparison FRQs like 2021 FRQ Q1 (hospital length of stay) and 2025 FRQ Q1 (gas mileage for cars from two countries) also lean on shape. When you compare distributions, use comparative language ("Country A's distribution is more strongly skewed right than Country B's") instead of describing each group separately.

Skew vs Skewed left vs. skewed right

The most common mistake is naming the skew after where the data piles up. It's the opposite. You name it after the tail. If most values cluster on the left and a long tail stretches to the right (toward bigger numbers), that's skewed right. Income and home prices are classic examples, since most values are moderate and a few are huge. If the pile is on the right and the tail stretches left, that's skewed left, like scores on an easy test where most people do well and a few score low. A quick check is to see which way the mean moves. It follows the tail.

Key things to remember about Skew

• Skew means a distribution is lopsided, with one tail stretching out much farther than the other.

• You name the skew after the direction of the long tail, so a tail toward larger values means skewed right.

• In a right-skewed distribution the mean is usually larger than the median, and in a left-skewed distribution the mean is usually smaller (EK 1.8.B.1).

• When a distribution is strongly skewed or has outliers, the median and IQR are better summaries than the mean and standard deviation because they are resistant.

• On a boxplot, a longer whisker or a median sitting off-center in the box signals skew toward the longer side.

• Always describe shape in context, and in Unit 4 check for strong skew before using t-procedures on small samples.

Frequently asked questions about Skew

What is skew in AP Statistics?

Skew describes a distribution that isn't symmetric because one tail is longer than the other. It's skewed right if the tail stretches toward larger values and skewed left if it stretches toward smaller values, and it's covered in Topic 1.8 under learning objective AP Stats 1.8.B.

Does skewed right mean most of the data is on the right?

No. Skewed right means the long tail is on the right, so most of the data is actually bunched on the left. You always name skew after the tail, not the peak.

Is the mean always greater than the median in a right-skewed distribution?

Not always, just usually. The CED (EK 1.8.B.1) says the mean is usually larger than the median for right skew, so on the exam you should treat it as a strong tendency rather than a guaranteed rule.

What's the difference between skew and an outlier?

Skew describes the overall shape of the whole distribution, while an outlier is a single value that falls unusually far from the rest. They often go together, since outliers tend to sit in the long tail, but a distribution can be skewed with no outliers. You can also have an outlier in an otherwise symmetric distribution.

Which measure of center should I use for skewed data?

Use the median, and pair it with the IQR for spread. Both are resistant to extreme values, so they give a more typical picture than the mean and standard deviation, which get pulled toward the tail.