Descriptive Statistics Calculator
Paste numbers to get mean, median, mode, standard deviation, variance, quartiles, IQR, standard error, skewness and kurtosis, with potential outliers flagged.
- ✓Mean, median, mode and range
- ✓Sample and population SD and variance
- ✓Quartiles, IQR and 1.5 × IQR outliers
Your statistics will appear here
Paste data on the left and click Calculate statistics
| Statistic | Value |
|---|---|
| Minimum | — |
| Q1 | — |
| Median | — |
| Q3 | — |
| Maximum | — |
Values outside these fences may deserve investigation. They are not automatically errors.
Descriptive Statistics: What the Numbers Mean
Use the free calculator above to summarize any numerical dataset. This guide covers what each statistic measures, how to read the results, and when to use sample versus population formulas.
What Are Descriptive Statistics?
Descriptive statistics summarize a set of observed values. They describe where numbers tend to sit (center), how widely they vary (spread), where particular values fall in the sorted order (position), and whether the distribution looks symmetric or leans to one side (shape).
This is different from inferential statistics, which use a sample to draw conclusions about a wider population. Descriptive statistics make no claim beyond the data you actually have. They do not test hypotheses, establish cause and effect, or determine whether a pattern is statistically significant.
This free descriptive statistics calculator takes a list of numbers and returns the complete picture in one go: mean, median, mode, variance, standard deviation, quartiles, IQR, standard error, potential outliers, skewness, and excess kurtosis. All calculations run in your browser. Nothing is uploaded.
How to Use This Calculator
- Paste your numbers into the data box. Commas, spaces, tabs, and line breaks all work as separators. Copying a single column from Excel or Google Sheets and pasting it directly is fine.
- Choose a quartile method under Advanced options if you need a specific convention. The default (Inclusive / linear interpolation) matches Excel's
QUARTILE.INCfunction. - Set decimal places if the default of four does not suit your reporting needs.
- Click Calculate statistics. Results appear on the right, grouped by category.
- Read the interpretation below the results for a plain-language summary of what the numbers show.
- Copy individual values by clicking the copy icon on each result card, or use Copy all statistics for a full plain-text export.
- Download a CSV when you need the results in a spreadsheet.
1234 rather than 1,234 because a comma separates observations. The calculator ignores blank entries and flags any non-numeric tokens so you can fix them before interpreting results.Worked Example
Click Example in the calculator to load this classic teaching dataset, which was chosen because it gives a clean set of results worth stepping through:
2, 4, 4, 4, 5, 5, 7, 9The mean is 40 ÷ 8 = 5. The median is the average of the 4th and 5th sorted values (4 and 5), giving 4.5. The value 4 occurs three times, so it is the only mode. The sum of squared deviations from the mean is 32. Dividing by 8 gives a population variance of 4, so population SD = √4 = 2. Dividing by 7 gives a sample variance of approximately 4.5714, and sample SD ≈ 2.1381.
These figures describe this specific dataset. They say nothing about any larger population from which the values might have been drawn.
Mean, Median and Mode
All three measure where values tend to cluster, but they respond differently to the shape of the data. A thorough explanation of each is in the guide to mean, median and mode.
Mean
The mean is the sum of all values divided by the count. It uses every observation, so a single extreme value can pull it noticeably away from the center. For the example above: 40 ÷ 8 = 5.
Median
Sort the values and take the middle one. For an even number of observations, average the two middle values. The median is not affected by extreme values the way the mean is, which makes it useful when the data contain notable high or low points.
Mode
The value (or values) that appear most often. A dataset can have one mode, several modes, or no mode at all. When every value is unique, this calculator returns "No mode" rather than listing all values. When two or more values tie for the highest frequency, all are returned.
Range, Variance and Standard Deviation
Spread statistics describe how far values tend to be from each other or from the center. For a deeper look at the most common one, see the guide to standard deviation.
Range
Maximum minus minimum. Quick to compute but only uses the two most extreme values, so it does not say much about where most of the data sits.
Variance
The average of the squared deviations from the mean. Squaring makes the units awkward (if your data are in centimetres, variance is in cm²), which is why standard deviation is usually preferred for reporting. Both sample and population versions are shown; see the next section for the distinction.
Standard Deviation
The square root of variance, which returns the result to the original units. A larger standard deviation means values are spread more widely from the mean; a smaller one means they cluster more tightly. What counts as "large" or "small" depends entirely on the measurement scale and context.
Sample vs Population Standard Deviation
This calculator shows both, labelled clearly, because the choice matters and is not always obvious.
| Version | Symbol | Denominator | Use when |
|---|---|---|---|
| Population | σ, σ² | N (all observations) | The values you entered are the entire group you want to describe |
| Sample | s, s² | n − 1 | The values are a sample drawn from a larger population you want to estimate variability for |
With just one observation (n = 1), sample variance is mathematically undefined. The calculator displays "n/a" in that case rather than showing a misleading zero.
Quartiles and IQR
Quartiles divide the sorted data into four equal parts. Q1 marks the boundary below which 25% of values fall; Q3 marks the boundary below which 75% fall. The interquartile range (IQR = Q3 − Q1) describes the spread of the middle 50% of the data.
QUARTILE.INC, its QUARTILE.EXC, and the median-of-halves method used in many statistics textbooks can all return different Q1 and Q3 for the same dataset. This calculator shows which method it used so you can compare results consistently. You can change the method under Advanced options.Five-Number Summary
The five-number summary (minimum, Q1, median, Q3, maximum) gives a compact picture of the distribution's center, spread, and range. It is the foundation of a box plot, which displays all five values visually. If you want to see the distribution's shape across value ranges, try the Histogram Maker.
Potential Outliers
The calculator uses Tukey's 1.5 × IQR rule to flag values that fall outside the fences Q1 − 1.5 × IQR (lower) and Q3 + 1.5 × IQR (upper). Values outside these fences are listed as potential outliers.
A flagged value is not automatically wrong. It could be a valid extreme observation, a measurement error, or a data-entry mistake. Check the original data before drawing any conclusion. Because the fences depend on Q1 and Q3, changing the quartile method may change which values are flagged.
Standard Error
Standard error (SE = s ÷ √n) estimates how precisely the sample mean estimates the true population mean. A smaller SE means the mean is likely to be a more stable estimate.
Standard error is not the same as standard deviation. Standard deviation describes how widely individual observations spread around the mean. Standard error describes how precisely the mean itself is estimated, and it shrinks as sample size grows even when the underlying variability stays the same.
Skewness and Excess Kurtosis
Skewness
Skewness measures the asymmetry of the distribution. This calculator uses the adjusted Fisher–Pearson sample skewness. A negative value suggests the tail extends more toward lower values; a positive value suggests the tail extends toward higher values. For small datasets, skewness estimates are unreliable on their own.
When all values are identical (zero variance), skewness is undefined and the calculator displays "n/a" rather than returning a meaningless number.
Excess Kurtosis
Excess kurtosis is the standard kurtosis value minus 3, so a normal distribution scores 0. Positive excess kurtosis indicates heavier tails relative to a normal distribution; negative indicates lighter tails. This calculator clearly labels the result Excess kurtosis to avoid ambiguity with definitions that do not subtract 3.
Like skewness, excess kurtosis requires at least four observations to be meaningful and is undefined when variance is zero.
Formulas Reference
All calculations use full internal precision. Rounding applies only to the displayed output.
Sum of all observations divided by the count.
| Statistic | Symbol | Formula | Notes |
|---|---|---|---|
| Population variance | σ² | Σ(xᵢ − μ)² / N | Divides by N |
| Sample variance | s² | Σ(xᵢ − x̄)² / (n−1) | Requires n ≥ 2 |
| Population SD | σ | √σ² | |
| Sample SD | s | √s² | Requires n ≥ 2 |
| Standard error | SE | s / √n | Requires n ≥ 2 |
| Range | Maximum − Minimum | ||
| IQR | Q3 − Q1 | Depends on quartile method | |
| Lower fence | Q1 − 1.5 × IQR | ||
| Upper fence | Q3 + 1.5 × IQR | ||
| CV | s / |x̄| × 100% | Undefined when mean = 0 or negative | |
| Skewness | Adjusted Fisher–Pearson (n ≥ 3, s > 0) | ||
| Excess kurtosis | Standard kurtosis − 3 (n ≥ 4, s > 0) |
Variance is computed using Welford's one-pass algorithm, which avoids the floating-point cancellation that can occur with large values close together.
Common Questions
Can I paste data from Excel or Google Sheets?
Yes. Copy a single column of numbers and paste directly into the data box. Tab-separated and newline-separated values are both handled.
Why do quartiles differ between calculators?
There is no single universally agreed method for computing quartiles. Excel's QUARTILE.INC, QUARTILE.EXC, and the median-of-halves method common in textbooks can all return different Q1 and Q3 from the same data. This calculator states its method clearly so you can match it to whatever source you are cross-checking against.
Can a dataset have more than one mode?
Yes. When two or more values tie for the highest frequency, all are returned. If no value repeats, the calculator shows "No mode" rather than listing every unique value.
Which standard deviation should I use?
Use population SD (σ) when the values you entered are the complete group you want to describe. Use sample SD (s) when the values are a sample drawn from a larger population and you want to estimate how variable that population is.
Is standard error the same as standard deviation?
No. Standard deviation describes how individual observations vary around the mean. Standard error describes how precisely the sample mean estimates the population mean. Standard error shrinks as sample size increases even when the data's underlying variability stays the same.
Should I remove potential outliers?
Not automatically. A value outside the 1.5 × IQR fences could be a valid extreme observation, a measurement error, or a data-entry mistake. Investigate before deciding anything.
Is my data uploaded anywhere?
No. All calculations run in JavaScript in your browser. The values you paste are not sent to any server.
What is the maximum number of values I can enter?
Up to 10,000 observations. The calculator will flag the input if the limit is exceeded.
What happens if I enter invalid text?
The calculator flags the problem and does not silently drop or convert invalid tokens. Fix the input and recalculate.
One calculation. The complete summary.
The free calculator is fastest when you have a list of numbers and want a clear, complete descriptive summary without building a full analysis workspace. It does not test hypotheses, prove significance, or choose a statistical method for you.
Descriptive Statistics Calculator questions
How the free browser tool works.
What input formats does the calculator accept?+
Numbers separated by commas, spaces, tabs or line breaks, including a column pasted from Excel or Google Sheets. Enter 1234, not 1,234, because commas separate values.
Does it show sample and population standard deviation?+
Yes. Both are always shown. Use population SD when your values are the whole group, and sample SD when they are drawn from a larger population.
Why might quartiles differ from Excel or my textbook?+
Quartile methods differ. The default matches Excel QUARTILE.INC, and you can switch to QUARTILE.EXC or the median-of-halves method. The method used is shown next to Q1 and Q3.
Is my data uploaded to a server?+
No. All calculations run in your browser.