Index / Mean, Median & Mode

Mean, Median & Mode Calculator

Paste a data set and get the complete descriptive summary — centre, spread, quartiles and a full frequency table, with the skew diagnosed for you.

Descriptive Summary

idle
n
Σx
min
max
skipped
Separate values with spaces, commas, semicolons, tabs or new lines. Anything that is not a number is skipped and reported.
Arithmetic mean
Median
Mode
Range
Midrange
Q1
Q3
IQR
SD (sample)
ValueFrequencyCumulativePercent

How to use it

Paste your values into the box. Separators do not matter — spaces, commas, semicolons, tabs and line breaks all work, so a column copied straight out of a spreadsheet pastes correctly without cleaning. Anything that does not parse as a number is skipped, and the count of skipped entries appears in the header strip. Always glance at that figure: a non-zero skip count usually means a stray label, a currency symbol or a thousands separator came along with the paste.

Results update as you type. The frequency table underneath lists every distinct value with its count, running cumulative total and percentage, with the modal value marked.

The three measures of centre

These three answer the same question — what is a typical value here? — and disagree in informative ways.

Arithmetic mean
x̄ = Σx ÷ n

The mean adds every value and divides by how many there are. It uses all the information in the data, which is its strength, and it is the foundation of variance, standard deviation, regression and most inferential methods. Its weakness is that every observation gets a vote weighted by its magnitude, so a single extreme value shifts it. Add one billionaire to a room of a hundred people and the mean wealth becomes a number describing nobody present.

The median is the middle value once the data is sorted — or the average of the two middle values when n is even. It cares only about rank order, so an extreme observation can move arbitrarily far away without shifting it at all. That robustness is why income, house prices, response times and hospital stay lengths are almost always reported as medians.

The mode is the most frequently occurring value. It is the least used of the three for continuous data, where exact repeats may not occur at all, but it is the only one that works on categorical data. There is no mean shoe size preference or median favourite colour; there is a modal one. A distribution can also have no mode, or several.

Reading the gap between mean and median

The relationship between the mean and median is a fast diagnostic for the shape of a distribution, and this calculator reports it directly.

PatternShapeTypical cause
mean ≈ medianSymmetricHeights, measurement error, test scores
mean > medianRight (positive) skewA long tail of high values — income, wait times, sales per customer
mean < medianLeft (negative) skewA tail of low values — exam scores near a ceiling, age at death

When the two diverge substantially, that divergence is a finding, not an inconvenience. Report both. A dataset where the mean is 42 and the median 28 is telling you something important about its tail that neither number conveys alone.

A worked example

Take the default data set: 12, 15, 15, 18, 22, 24, 24, 24, 27, 31, 33, 40 — twelve values.

  • Sum = 12+15+15+18+22+24+24+24+27+31+33+40 = 285
  • Mean = 285 ÷ 12 = 23.75
  • Median: n is even, so average the 6th and 7th sorted values — (24 + 24) ÷ 2 = 24
  • Mode: 24 appears three times, more than any other value — 24
  • Range = 40 − 12 = 28

Mean 23.75 against median 24 is a difference of a quarter of a unit on data spread across 28 — near-perfect symmetry. The single value of 40 in the upper tail is not extreme enough to drag the mean meaningfully. Compare what happens if you change that 40 to 400: the median stays at 24, while the mean jumps to 53.75, a figure larger than every observation except one. That is the entire argument for the median in one edit.

Quartiles and the interquartile range

The calculator also reports Q1, Q3 and the IQR, because they describe spread in the same robust spirit as the median. Q1 is the value below which a quarter of the data falls; Q3 the value below which three quarters falls. The IQR is the distance between them — the width of the middle half of the data.

Interquartile range
IQR = Q3 − Q1

Outlier fences: Q1 − 1.5×IQR and Q3 + 1.5×IQR

Those fences are the standard convention behind box plot whiskers. A point beyond them is conventionally flagged for inspection — not deleted. An outlier is a question, not a defect: it may be a transcription error, or it may be the most interesting observation in the set.

On quartile conventions. There is no single agreed method. This tool uses linear interpolation between order statistics, matching R's type 7 default and Excel's PERCENTILE.INC. Minitab, some textbooks and Excel's PERCENTILE.EXC use different rules. On large datasets the difference is invisible; on ten observations it can shift Q1 by a noticeable amount. If your answer must match a specific textbook, check which convention it uses.

Which measure should you report?

  • Roughly symmetric, no extreme values — the mean, and it will be close to the median anyway.
  • Skewed or containing outliers — the median, ideally alongside the IQR rather than the standard deviation.
  • Categorical or ordinal data — the mode, since neither of the others is defined.
  • You need to do further statistics — the mean, because variance, standard deviation, confidence intervals and regression are all built on it. The median is descriptively better but analytically harder to work with.
  • When unsure — report both and let the reader see the gap. That is more honest than picking the one that suits your argument.

Frequently asked questions

Can a data set have more than one mode?
Yes. If two or more values tie for the highest frequency, the distribution is bimodal or multimodal and this tool lists them all. Genuine multimodality is worth investigating — it often means you have two different populations mixed together, such as measurements from two machines or two age groups pooled into one column.
What if no value repeats?
Then there is no mode, and the tool reports "none". This is normal for continuous measurements recorded to any precision. Some textbooks say every value is a mode in that case, which is technically defensible and practically useless; we report none.
Why is my median not one of my data values?
Because n is even. With an even count the median is the average of the two middle values, which need not be a value that actually occurs. With 12 observations the median is the mean of the 6th and 7th.
Does the order I paste values in matter?
No. The tool sorts internally before computing the median and quartiles, so any order gives the same result. The frequency table is always displayed in ascending value order regardless of input order.
How many values can I paste?
Up to 100,000. Beyond a few thousand the frequency table becomes long, but the summary statistics remain instant. Everything is computed in your browser, so performance depends on your device rather than on any server.
Should I use this for grouped or frequency data?
Not directly — it expects raw values. If you have a frequency table, expand it first, or use a weighted mean instead. A value of 24 with frequency 3 should appear as three separate 24s in the paste.
Is my data stored?
No. Everything runs in your browser. Nothing is transmitted to a server, logged or saved, and closing the tab discards it entirely.

Next steps

Once you know the centre of your data, the natural next question is how spread out it is — that is the standard deviation calculator. With a mean and a standard deviation in hand you can then ask how unusual any single observation is using the z-score calculator, or how precise your estimate of the mean actually is with the confidence interval calculator.