Mean or Median? Picking the Right Average for Your Data

Average is a loose word covering at least three different calculations, and the one you pick changes the story the number tells. That is not a technicality. It is the reason two people can quote honest figures about the same dataset and disagree completely.

One dataset, two very different answers

Take seven salaries at a small firm: 32,000, 35,000, 36,000, 38,000, 41,000, 44,000 and 250,000. The mean is the total divided by the count, so 476,000 divided by 7 gives 68,000. The median is the middle value once they are sorted, which is 38,000.

Every one of those figures is correct. But six of the seven people earn less than the mean, and describing 68,000 as a typical salary at that firm would be misleading to the point of dishonesty. The median sits where a real employee actually sits. That gap is exactly what a single large value does to a mean, and it is why national income statistics are almost always reported as medians.

Where the mean earns its keep

Symmetric data is the mean's home ground. Take 4, 4, 6, 7 and 9: the mean is 30 divided by 5, which is 6, and the median is also 6. When the two agree closely, the distribution is roughly balanced and the mean is the better summary because it uses every value rather than just the middle one.

The mean is also the only one of the three that behaves under arithmetic. Multiply it by the count and the total comes back, which matters for budgets. Medians do not work that way, so payroll needs the mean even when a journalist needs the median.

What mode and range are actually for

The mode is the most frequently occurring value: in that 4, 4, 6, 7, 9 set it is 4. It is the only measure of centre that works on data with no numeric order, such as the most common shoe size sold or the most popular delivery day. On continuous data like heights or salaries, the mode is often useless because no two values repeat exactly.

Range is fragile

Range is the largest value minus the smallest. For the salary set, 250,000 minus 32,000 gives a range of 218,000, which is a number built entirely from the two most unusual people in the room. Change one of them and the range changes completely; change anyone in the middle and it does not move at all. It is a quick sanity check on spread, not a serious measure of it. For that, standard deviation is the tool, because it accounts for every value's distance from the mean.

Using the mean median mode range calculator

Paste your numbers into the input box separated by commas, spaces or line breaks. The calculator sorts them for you, so there is no need to put them in order first, and it counts the values as it goes, which catches the common mistake of pasting one number short.

Copy from a spreadsheet column rather than retyping. Watch for two problems: thousand separators such as 1,200 can be read as two values, 1 and 200, and currency symbols may break the parse. Strip the formatting first, then check the reported count against the rows you copied.

Read all four outputs together rather than picking one. If the mean and median are close, either summarises the data fine. If they are far apart, you have skew, and the direction tells you which tail is long. A step-by-step walkthrough of each formula is in our mean, median and mode calculator guide.

Common questions

Can a dataset have more than one mode? Yes. Two values tied for most frequent makes it bimodal, and more than two makes it multimodal. If every value appears exactly once, most conventions say there is no mode at all rather than saying all of them are.

How do I find the median with an even number of values? Sort them and average the two middle ones. With eight values, add the fourth and fifth and divide by two. The result may not be a number that appears in the set, which is fine.

Which average do house prices use? The median, almost always. A handful of very expensive sales would pull a mean upwards and make the typical home look pricier than it is.

Does removing outliers fix a skewed mean? It changes the mean, but deciding what to delete is a judgement call you have to justify. Reporting the median alongside the mean is usually more honest than dropping inconvenient rows.