Written by Sarah Mitchell · Reviewed by Jane Smith · August 1, 2026
Compute descriptive statistics — mean, median, mode, variance, standard deviation, and more.
Enter data and click Calculate.
Averages get quoted everywhere, but a lone mean can hide a lot. This page computes a full battery of descriptive statistics at once, so you can see the center of your data and its spread side by side instead of trusting one number. Because everything runs in the browser, you can paste in sensitive grade sheets or lab results without worrying about them leaving the machine.
Suppose a class of ten students scores 42, 55, 61, 61, 67, 72, 78, 78, 78, and 90 on a quiz. Adding all ten values gives 682, so the mean is 682 ÷ 10 = 68.2. To find the median, sort the list and take the middle: with an even count, the median is the average of the fifth and sixth scores, (67 + 72) ÷ 2 = 69.5. The mode is 78, because it appears three times and no other score repeats that often. The range is 90 − 42 = 48, which shows the papers spanned almost half the quiz.
Two data sets can share the same mean yet look completely different. The variance and standard deviation measure how tightly the values cluster around the center. For the ten scores above, the population standard deviation is about 13.2, which reflects the wide gap between the lowest and highest papers. A small standard deviation means most values sit close to the mean; a large one means the data is scattered.
Income, rent, and real-estate data are classic cases where the mean gets dragged upward by a few very large values. In that situation the median gives a fairer picture of the typical record, because it only cares about the middle of the sorted list. The mode is most useful for categorical-style questions, like finding the most common delivery time or the best-selling shoe size, where a numeric midpoint does not really exist.
The input box accepts numbers separated by commas, spaces, tabs, or line breaks, and you can mix those separators freely in one paste. Non-numeric tokens are skipped rather than treated as errors, so stray labels in a copied column will not abort the run. Decimal values, negative numbers, and zero are all valid inputs wherever the math allows them.
The results are grouped so the headline measures appear first, followed by the spread measures. Count and sum confirm that the calculator saw exactly the values you intended, which is the fastest way to catch a missing row or an accidental extra number. If no value repeats, the mode reports "No mode" instead of inventing one, and bimodal data lists every value tied for the top frequency.
Use the population formulas when your list contains every member of the group you care about, such as all ten quiz scores in a single class. Use the sample formulas when your list is a small slice drawn from a larger population, because dividing by N−1 nudges the variance up and removes the bias that small samples would otherwise carry.
Skew describes which tail of the distribution is longer. Income figures are the textbook example of right-skewed data: a handful of very large salaries stretch the right tail, and because the mean is sensitive to every value, it climbs above the median. A left-skewed set behaves the opposite way, dragging the mean below the median. Comparing the two numbers rather than reading one of them in isolation is the fastest diagnostic for skew.
Results are reported to four decimal places when the raw value carries more digits, which preserves intermediate precision for anyone feeding the numbers into a follow-up calculation. For a report or a dashboard, rounding to two decimals is usually enough and reads more cleanly. If you round during the calculation instead of at the end, the final answer can drift by a cent or two, so the safer habit is to round only the displayed result.
Percentiles describe where a value sits inside the distribution. The 25th percentile is the point below which a quarter of the data falls, the 50th is the median, and the 75th marks the three-quarter line. The gap between the 75th and 25th percentiles is the interquartile range, a robust measure of spread that ignores extreme outliers entirely. When you compare two groups, a box plot built from these five numbers often tells the story faster than a table of means, because it exposes skew, symmetry, and gaps between clusters in a single picture.