Choosing the right numerical summary matters because unusual observations can distort some measures much more than others. AP Statistics expects you to recognize which summaries stay stable and which are easily pulled.

Understanding resistance

A summary measure can be described by how strongly it reacts when a data set includes an unusually large or unusually small value. This matters because real data often contain a few observations that are far from the rest.

Resistance means stability, not complete immunity.

In AP Statistics, the key distinction is that the median and IQR are resistant, while the mean, standard deviation, and range are nonresistant. When you choose a summary, you are deciding whether protection from the influence of outliers is important.

Why some measures are more sensitive

Mean and standard deviation

The mean uses every observation in the data set. Because each value contributes to the total, a single extreme observation can pull the mean upward or downward. The larger that unusual value is, the more influence it can have.

The standard deviation is also nonresistant because it measures spread around the mean. If one observation lies far from the rest, the overall variability increases, and the standard deviation can become much larger.

This means an outlier can affect both the reported center and the reported spread when these measures are used.

Range

The range is especially sensitive because it depends only on the smallest and largest observations. If either endpoint changes, the range changes immediately. One extreme value can therefore make the data appear much more spread out than most observations actually suggest.

Median and IQR

The median depends on position in an ordered list, not on the exact size of the most extreme values. The IQR is based on the middle half of the data, so it usually ignores unusually small or large observations near the ends. As a result, these two measures give a more stable picture of center and spread when outliers are present.

This figure illustrates a standard boxplot, where the median is marked inside the box and the IQR is represented by the box’s width (from $Q_1$ to $Q_3$). Points beyond the whiskers are plotted as outliers, visually emphasizing that median and IQR are determined by position and the middle portion of the data rather than the most extreme values. Source

Choosing appropriate summaries

When a data set contains influential outliers, median and IQR are usually the better choice. They describe the typical value and the variability of the bulk of the data without allowing a few unusual observations to dominate the summary.

When outliers are not strongly affecting the data, mean and standard deviation may still be useful because they incorporate all observations. However, once extreme values become influential, these nonresistant measures can give a misleading impression of both center and spread.

A strong AP Statistics response should not only name a measure but also explain why it is appropriate. The essential reasoning is:

• if outliers are influential, choose median and IQR because they are resistant

• avoid relying on mean, standard deviation, and range because they are nonresistant

• if you report center and spread together, pair measures with similar behavior toward outliers

One common pairing is median with IQR. Another is mean with standard deviation. These pairings are helpful because the measures in each pair respond to unusual values in a consistent way. The range can still describe overall extent, but it is rarely the best main measure of spread when resistance matters.

What resistance means in practice

Resistant measures do not ignore the data set. Instead, they prevent a small number of extreme observations from controlling the summary. This is important when the goal is to describe what is typical for most individuals in the data.

Nonresistant measures are not bad measures. They are simply more sensitive. That sensitivity can be helpful when every observation should strongly affect the summary, but it can also be a weakness when a few extreme values are not representative of the broader pattern.

This is why AP Statistics often expects you to think about the role of outliers before selecting summary statistics. The choice is not automatic; it depends on whether you want a summary that stays stable or one that reacts strongly to extremes.

Common decision rules

Questions to ask before choosing

Before selecting a summary, ask:

• Are there observations much larger or smaller than most of the data?

• Would one extreme value change the reported center a lot?

• Do you want to describe the middle of the data or the full impact of every value?

If the answer to the first two questions is yes, a resistant summary is usually more appropriate. If not, a nonresistant summary may be acceptable.

Common mistakes to avoid

Students often lose points by identifying an outlier but still choosing the mean or standard deviation without justification. Another common error is naming the median as resistant but pairing it with the range, which is not resistant. It is also not enough to say that one measure is better without stating the reason. The reason must be connected to the influence, or lack of influence, of outliers.

When you justify a choice, use language such as "the median is resistant to outliers" or "the standard deviation is nonresistant and can be inflated by extreme values." Clear wording shows that you understand the statistical property, not just the vocabulary.

Comparing resistant and nonresistant perspectives

In some contexts, both types of summaries may be reported, but they emphasize different features of the data. Median and IQR focus on the middle portion of the distribution, while mean, standard deviation, and range are more affected by unusual values. The key AP skill is recognizing which set of measures better represents the data when outliers are present.