Descriptive Statistics Calculator
Paste a column of numbers to describe it: the middle, the spread…
Paste a column of numbers to describe it: the middle, the spread, the quartiles by the method your class or software uses, and any outliers, drawn as a box plot.
Read 25 values; left out the header row (“Commute (minutes)”).
ExampleMedian, middle half
33
Half the values lie between 29 and 39 (IQR 10). 1 value is an outlier beyond 1.5 × IQR: 95.
Paste your own column of numbers.
The histogram counts values in equal bins; under it the box is the middle half, the line the median, the diamond the mean, and dots beyond the dashed fences are outliers.
- Mean (SD)
- 35.84 (13.86)
- Five-number summary
- 22 · 29 · 33 · 39 · 95
- Skewness
- 3.409 (long right tail)
How this works: 9 sources,
How this works
Method
Descriptive statistics summarise a list of numbers before any test: where the middle is, how spread out the values are, and whether any stand apart. The five-number summary (minimum, first quartile, median, third quartile, maximum) and the box plot drawn from it show all of that at a glance. In the example, 25 commute times have a median of 33 minutes and one 95-minute trip that is far from the rest. Full method
Sources
- NIST/SEMATECH e-Handbook of Statistical Methods, 1.3.3.7 Box Plot
- Hyndman and Fan, Sample quantiles in statistical packages (The American Statistician, 1996)
- NIST/SEMATECH e-Handbook of Statistical Methods, 7.2.6.2 Percentiles
- NIST/SEMATECH e-Handbook of Statistical Methods
- Abramowitz and Stegun, Handbook of Mathematical Functions, chapters 6, 7 and 26 (1964)
- SciPy 1.18 scipy.stats (reference values for the tests)
- statsmodels 0.15 (reference values for the A/B test and regression intervals)
- Welch, The generalization of Student’s problem when several different population variances are involved (Biometrika) (1947)
- Fleiss, Levin and Paik, Statistical Methods for Rates and Proportions, 3rd edition (sample size for two proportions) (2003)
How it’s tested
3 worked examples for this page are checked by automated tests before every release: given the inputs, the tool must show the expected answer.
Privacy
On your device. Nothing you enter or open leaves this browser. About privacy labels
Changes
- First published: live curves with draggable cut-offs, and tests on pasted data (t-tests, ANOVA, chi-square, Fisher’s exact, correlation, regression, A/B), checked against SciPy.
How quartiles, the IQR and outliers are found
Descriptive statistics summarise a list of numbers before any test: where the middle is, how spread out the values are, and whether any stand apart. The five-number summary (minimum, first quartile, median, third quartile, maximum) and the box plot drawn from it show all of that at a glance. In the example, 25 commute times have a median of 33 minutes and one 95-minute trip that is far from the rest.
Quartiles split the sorted data into four parts, but there is no single rule for where the split falls between two values, and software disagrees. Excel’s QUARTILE.INC, Google Sheets, R and NumPy interpolate at position (n − 1)p + 1; Excel’s QUARTILE.EXC, Minitab and SPSS use (n + 1)p; TI-84 calculators and most school textbooks take the median of each half. Hyndman and Fan catalogued nine definitions in 1996. Pick the one your course or software uses and the numbers will match.
The interquartile range, Q3 − Q1, is the spread of the middle half. John Tukey’s fences sit 1.5 IQR below Q1 and above Q3; values beyond them are flagged as outliers, and beyond 3 IQR as far out. An outlier is a value to check, not one to delete: it may be a typing error, or the most interesting number in the data. The mean is pulled towards it; the median is not, which is why a long right tail shows as a mean above the median and a positive skewness.
IQR = Q3 − Q1 · fences = Q1 − 1.5 IQR and Q3 + 1.5 IQR
- A box plot shows the median, the lower and upper quartiles, and the minimum and maximum, with points beyond 1.5 × IQR often drawn separately as possible outliers. Source: NIST/SEMATECH e-Handbook of Statistical Methods, 1.3.3.7 Box Plot.
- Hyndman and Fan (1996) described nine definitions of sample quantiles used by statistical packages and recommended one of them (type 8); R’s default is type 7. Source: Hyndman and Fan, Sample quantiles in statistical packages (The American Statistician, 1996).
- The e-Handbook notes that there is no universal agreement on how to compute percentiles and gives the (N + 1)p definition alongside others. Source: NIST/SEMATECH e-Handbook of Statistical Methods, 7.2.6.2 Percentiles.
The three quartile methods side by side
Q1 and Q3 of the numbers 1 to n by each method, showing how far they can disagree on small samples (they converge as n grows).
| Data | Excel, Sheets, R | Excel EXC, Minitab | TI-84, textbooks |
|---|---|---|---|
| 1 to 4 | 1.75 and 3.25 | 1.25 and 3.75 | 1.5 and 3.5 |
| 1 to 5 | 2 and 4 | 1.5 and 4.5 | 1.5 and 4.5 |
| 1 to 6 | 2.25 and 4.75 | 1.75 and 5.25 | 2 and 5 |
| 1 to 7 | 2.5 and 5.5 | 2 and 6 | 2 and 6 |
| 1 to 8 | 2.75 and 6.25 | 2.25 and 6.75 | 2.5 and 6.5 |
| 1 to 9 | 3 and 7 | 2.5 and 7.5 | 2.5 and 7.5 |
| 1 to 10 | 3.25 and 7.75 | 2.75 and 8.25 | 3 and 8 |
| 1 to 11 | 3.5 and 8.5 | 3 and 9 | 3 and 9 |
| 1 to 12 | 3.75 and 9.25 | 3.25 and 9.75 | 3.5 and 9.5 |
| 1 to 15 | 4.5 and 11.5 | 4 and 12 | 4 and 12 |
Show all 11 rowsShow fewer
| 1 to 20 | 5.75 and 15.25 | 5.25 and 15.75 | 5.5 and 15.5 |
Common questions
Why do my quartiles differ from my calculator’s?
Different tools use different definitions. A TI-84 takes the median of each half; Excel’s QUARTILE.INC interpolates. Switch “Quartiles as in” to match your tool; the worked example shows all three.
Should I remove outliers?
Only with a reason: a measurement or typing error, or a value from outside the population you are studying. Otherwise keep them and report results with and without, or use methods based on ranks or medians that they barely move.
Where are the mean, median and mode worked out step by step?
In the Math Calculator’s mean, median and mode page, which shows every step with exact fractions. This page focuses on the shape of the data: quartiles, spread, outliers and the histogram.