The Three, Side by Side

MeasureHow You Get ItReach For It When...
MeanAdd everything up, divide by how many values there areThe numbers are reasonably close together, no wild outliers
MedianLine the values up and pick the one in the middleA few extreme values would otherwise throw things off
ModeWhatever value shows up most oftenYou care about the most common result, not a calculated one

Calculating the Mean

This is the version most people learn first and just call "the average." Total up every value in the set, then divide by however many values you added.

Formula: Mean = (Sum of all values) ÷ (Number of values)

Test scores: 72, 85, 91, 68, 88, 76
Sum = 72 + 85 + 91 + 68 + 88 + 76 = 480
Count = 6
Mean = 480 ÷ 6 = 80

The weak spot: the mean gets pulled toward extreme values. Replace that 72 with a 20 and the mean drops to roughly 71 - a real shift caused by changing a single entry. That's the core limitation to remember about it.

Calculating the Median

Sort your values from smallest to largest, then pick the one that lands in the middle. With an even number of values there's no single center, so you average the two closest to it.

When you have an odd count:
Values: 3, 7, 8, 12, 15
Sorted: 3, 7, 8, 12, 15
Median = 8

When you have an even count:
Values: 4, 9, 13, 20, 25, 31
Middle two: 13 and 20
Median = (13 + 20) ÷ 2 = 16.5

Reach for the median whenever a small number of outliers could skew the picture. Salaries are the classic case - a handful of executives earning enormous sums can pull the mean well above what a typical worker actually takes home. The median doesn't budge no matter how far above or below the middle those extreme values sit, which makes it a more honest read on "typical."

Calculating the Mode

No math required here, just observation - find whichever value appears most frequently in your dataset.

Dataset: 5, 8, 3, 5, 9, 2, 5, 7, 8
5 shows up 3 times (more than anything else)
Mode = 5

A dataset can have more than one mode - two values tied for the top spot makes it bimodal, three or more makes it multimodal. It's also possible for a dataset to have no mode at all, when every value appears exactly once.

Mode is most valuable with categorical data, where mean and median simply don't apply. A shoe retailer can't order inventory based on a "mean shoe size," since that figure might not correspond to any size that actually exists. Knowing which size sells most often (the mode) tells them precisely what to stock.

Comparing All Three on One Dataset

The cleanest way to see how differently these three behave is to calculate all of them from a single set of numbers.

Monthly household income for 8 households, in dollars: 2400, 2800, 3100, 3100, 3600, 4200, 4800, 18000

Sorted: 2400, 2800, 3100, 3100, 3600, 4200, 4800, 18000

Mean = (2400+2800+3100+3100+3600+4200+4800+18000) ÷ 8
= 42000 ÷ 8 = $5,250

Median = (3100 + 3600) ÷ 2 = $3,350

Mode = $3,100 (appears twice)

Look at how far apart those three results are. A single household earning $18,000 dragged the mean all the way up to $5,250, overstating what most households in this group actually bring in. The median ($3,350) and mode ($3,100) land much closer to the typical figure. This is precisely why government statistics report median household income rather than the mean - the mean would make things look better than they are.

Matching the Right Measure to the Situation

SituationBetter ChoiceReasoning
Test scores in a classMeanScores usually bunch fairly close to a central value
House prices in a cityMedianA handful of luxury sales would inflate the mean
Most popular T-shirt sizeModeYou need the most common size, not a computed one
Average daily temperatureMeanTemperatures tend to vary smoothly and evenly
Salary comparisonsMedianA few very high earners create outliers
Most ordered pizza toppingModeCategorical data - a mean here wouldn't mean anything

Bonus: What Range Tells You

Range doesn't technically belong to the mean/median/mode trio, but it's usually taught alongside them because it captures something the other three miss entirely. It's just the gap between the highest and lowest values in a set, and it gives a quick sense of how spread out the data is.

Range = Maximum − Minimum

Values: 12, 15, 22, 31, 45
Range = 45 − 12 = 33

A large range points to scattered data; a small one points to values clustered tightly together. Pair the range with the mean and you get a much fuller picture than the mean alone could ever provide.

Skip the Manual Sorting

If you'd rather not sort and add everything by hand, SolverCalc's statistics calculator works out mean, median, mode, range, and standard deviation for any list you feed it. Type in the values separated by commas and let it do the rest.

The Bottom Line

Mean, median, and mode each measure something distinct, and none of them beats the others as a general rule - it always comes down to the data in front of you. The mean is the balance point of the entire set. The median is whatever sits in the middle once everything's sorted. The mode is simply what appears most. When there aren't extreme outliers, the mean does the job fine. But for income, home prices, ages, or anything else where a few extreme values could distort the picture, the median tends to give you the more accurate read. And whenever frequency matters more than magnitude, mode is the tool to reach for. Knowing which one applies where makes you a lot harder to mislead with statistics, and that's worth having in your back pocket.