How to find the mode (and when it is useful)

How to find the mode

The mode is the value that occurs most often in a data set. Finding it is counting, not arithmetic, so there is no formula to memorize.

1. Sort the values, or tally them, so identical values sit together.
2. Count how many times each distinct value appears.
3. The value with the largest count is the mode.
4. If two or more values tie for the largest count, every one of them is a mode.
5. If every value appears exactly once, the data set has no mode.

Sorting first is the part people skip, and it is where errors come from. In an unsorted list of 20 numbers it is easy to miscount a repeat, so write out the tally before you answer.

Finding the mode in a list

Take the number of bikes available at a share dock at 8 a.m. on 11 weekdays: 4, 7, 5, 7, 9, 7, 6, 5, 8, 7, 5. Sorted, that is 4, 5, 5, 5, 6, 7, 7, 7, 7, 8, 9.

Now the counts are easy to read: 4 appears once, 5 appears three times, 6 once, 7 four times, 8 once, and 9 once. Those counts add to $1 + 3 + 1 + 4 + 1 + 1 = 11$, which matches the sample size, so nothing was dropped. The biggest count belongs to the value 7, which shows up four times, so the mode is 7 bikes.

That check on the counts is worth doing every time. If your tallies do not add up to $n$, you have either missed a value or counted one twice.

Finding the mode in a frequency table

A frequency table has already done the counting for you. Scan the frequency column for the largest number, then read across to the category or value in that row.

The largest frequency is 24, so the mode is Navy. The mode is the color, not the number 24. Reporting 24 as the mode is the single most common error on frequency-table questions, and it is the same mistake as saying the tallest bar's height instead of what the bar stands for.

Relative frequencies point to the same answer, since dividing every count by the same total of $24 + 18 + 12 + 6 = 60$ does not change which row is largest. Here $\frac{24}{60} = 0.40$ is the largest proportion, and Navy is again the mode. For the difference between the two columns, see frequency vs relative frequency.

Finding the mode in a histogram: report the modal class

A histogram groups values into bins, so the raw values are no longer visible. You cannot recover the single most frequent value from it. What you report instead is the modal class, the bin with the tallest bar.

So if the tallest bar covers 60 to 70 minutes, the correct answer is "the modal class is 60 to 70 minutes," not a single number. Naming a specific value inside that bin claims information the graph does not contain.

There is a further catch. Changing the bin widths changes the appearance of a histogram, and it can change which bin is tallest, so the modal class depends partly on how the graph was drawn. A dotplot avoids the issue by showing every value, which is why it is the better graph for reading modes in a small data set. Compare the options in dotplot vs histogram vs stemplot.

No mode, one mode, or several

Unlike the mean and the median, the mode is not guaranteed to exist or to be unique.

• No mode. In 3, 5, 8, 11, 14 every value appears once, so there is no mode. Saying "all of them are modes" is not the usual convention; say the data set has no mode.
• One mode. In 2, 4, 4, 4, 7, 9 the value 4 appears three times and everything else appears once, so the mode is 4.
• Two modes. In 1, 1, 5, 5, 8 both 1 and 5 appear twice, so both are modes.

Be careful with one piece of vocabulary here. A data set with two tied modal values is not automatically a bimodal distribution. Bimodal describes a graph with two prominent peaks separated by a dip, so 1, 1, 2, 2, 3 has two modes but only one hump. Two modes far apart with a gap between them is the case that really is bimodal, and what a bimodal distribution means covers what to do about it.

The mode is the only center that works for nominal categorical data

This is the reason the mode earns its place. You cannot average colors, brands, blood types, or bus routes, and none of them has a meaningful order you could use to find a middle. The mean needs arithmetic and the median needs an ordering, so both are undefined for a nominal categorical variable, one whose categories have no natural order.

Ordinal categorical variables are the exception worth knowing. Letter grades, Likert responses running from strongly disagree to strongly agree, and sizes small, medium, large all carry a genuine order, so you can sort the responses and report a median category, and researchers routinely do. The mean is still out, because the gaps between neighboring categories are not measured quantities.

Counting still works either way, though. The most frequent category is always well defined, so the mode is the only measure of center that survives. When a news story says the most common major at a university, or the most popular first name in a birth year, that is a mode doing work no other summary could do.

The AP course reaches the same information under different names. Its Unit 1 treatment of one categorical variable is built on frequency tables, relative frequencies, proportions, and percentages, and on bar graphs where the tallest bar shows the most common category. That tallest bar is the mode, whether or not the word appears. See qualitative vs quantitative data if you are unsure which type you are holding.

Why AP Statistics leans on the mean and the median instead

For quantitative data, the Fall 2026 framework names two commonly used measures of center: the mean and the median. The word mode does not appear in the course framework at all, though the related shape vocabulary, unimodal and bimodal, does.

There are good statistical reasons behind that choice.

• The mode ignores almost all of the data. It reports one count and throws away the rest, while the mean $\bar{x}$ (read x-bar) uses every value.
• It is unstable. In the bike-dock list from earlier, changing the single 4 into a 5 makes 5 appear four times, tying 7, so the data set goes from one mode to two while the median stays at 7.
• It can sit far from the center. In 1, 1, 40, 50, 60 the mode is 1, which describes the low end rather than the middle.
• Nothing later depends on it. The standard deviation, z-scores, confidence intervals, and every test in Units 3 through 5 are built on means and proportions, not modes.

So report the mode when the question asks for the most common value or when the variable is categorical. For quantitative data, use the median for skewed distributions and the mean for roughly symmetric ones, as mean vs median works through in detail. You can also look up the term in the mode glossary entry.

Common mistakes

• Reporting the frequency instead of the value. If Navy was bought 24 times, the mode is Navy, not 24.
• Reading a single value off a histogram. Report the modal class, the tallest bin's interval.
• Calling a data set bimodal because two values tie. Bimodal describes two prominent peaks in the graph, not a tie in the counts.
• Answering "no mode" when there is a tie. Ties mean several modes; no mode means every value appears exactly once.
• Forgetting to sort or tally, then miscounting a repeat in a long list.
• Trying to find a mean or a median for a nominal categorical variable. Neither exists, so the mode is the answer you want. An ordinal variable does allow a median category.