Mean, standard deviation, median and more — 100% local.
Calculation performed entirely in your browser. The "population" variance and standard deviation divide by n, the "sample" version (used to estimate a population from a sample) divides by n−1.
The mean adds up all the values and divides by their count. The median is the middle value once the series is sorted. The difference matters as soon as there are extreme values: for the salaries 3000, 3200, 3500, 3800, 40000, the mean is 10,700 — a figure no one earns — while the median, 3500, describes the centre of the group well. The median resists outliers; the mean does not.
Standard deviation measures the spread around the mean. Two formulas coexist depending on what your series represents.
Divide by n if your series is the whole population studied (the marks of this class). Divide by n − 1 if it's only a sample meant to represent a larger whole (a survey). The "−1" corrects a bias: a sample naturally underestimates the true spread.
The quartiles Q1, Q2 (the median) and Q3 cut the sorted series into four equal parts. The box runs from Q1 to Q3 and therefore holds the central half of the data; its width (Q3 − Q1) is the interquartile range. The whiskers stretch to the values still considered "normal", and points beyond them are flagged as outliers.
Spaces, commas, semicolons and line breaks, freely mixed. The decimal separator is the point.
No. Everything is computed locally; the series you enter doesn't leave the browser and disappears when the tab is closed.
It's the average of the two central values of the sorted series.