Removed as outliers
Distribution (cleaned data)
This calculator extracts numbers from pasted text — labels, commas, and other characters are ignored automatically. Outliers are identified using the IQR method: values that fall more than a set multiple of the interquartile range beyond the first or third quartile are excluded, and every statistic below is calculated on the remaining, cleaned dataset.
The interquartile range (IQR) is the spread between the first quartile (Q1, the value below which 25% of the data falls) and the third quartile (Q3, below which 75% falls). A value is flagged as an outlier if it sits more than a chosen multiple of the IQR below Q1 or above Q3. This is Tukey's method — the same test used in standard box-and-whisker plots — and it doesn't assume the data is normally distributed, unlike methods based on standard deviations from the mean.
The slider sets the IQR multiplier used in the outlier test. The standard value, 1.5, is Tukey's original rule of thumb and works well for most everyday data. Lower values (down to 1.0) remove more aggressively, flagging borderline points as outliers too — useful when you want a very clean dataset. Higher values (up to 5.0) only catch extreme cases, useful when the data is naturally spread out and you don't want genuine variation mistaken for an error.
The parser pulls every number it finds out of whatever you paste — a spreadsheet column, a list with units or labels attached, even a sentence with figures embedded in it. Anything that isn't a number (commas, currency symbols, words) is simply ignored, so there's no need to clean the data up first.