Trimmed Mean Calculator
This calculator finds the trimmed mean of a data set, a robust average that reduces the influence of outliers by discarding a percentage of the highest and lowest values before averaging the rest. The ordinary mean is easily distorted by extreme values: a single very large or very small number can drag it well away from where most of the data sits. The trimmed mean guards against this by removing a set percentage from each end of the sorted data and taking the mean of what remains. It keeps most of the efficiency of the ordinary mean while gaining much of the robustness of the median, which is why it is used in fields from sports judging, where the highest and lowest scores are dropped, to economics and any analysis where a few extreme readings should not dominate. This calculator does it. You paste your data and choose the trim percentage to remove from each end, and it returns the trimmed mean, the ordinary mean for comparison, the number of values removed from each end, and the count remaining. The results update as you type. Use it for statistics study, for analysing data with outliers, or whenever you want a central value that is not skewed by extremes. The calculator sorts the data, removes the chosen percentage of values from the bottom and the top, then averages those that are left. A 10 percent trimmed mean, for example, discards the lowest 10 percent and the highest 10 percent of the data. The larger the trim percentage, the more robust the result but the more data you discard, with a trim of 50 percent converging on the median. Comparing the trimmed mean with the ordinary mean is itself informative: a large gap between them signals that outliers are pulling the ordinary mean around, which is exactly when the trimmed mean is the better summary.
Sorts the data, removes the trim % from each end, then averages the rest. More trim is more robust but discards more data. 50% trim approaches the median.
How it works
The data is sorted in order. The trim percentage is applied to the count to find how many values to remove from each end, rounded down, and those lowest and highest values are discarded. The trimmed mean is the average of the values that remain, reducing the pull of outliers compared with the ordinary mean.
Worked example
For the data 1 to 9 plus an outlier of 100, the ordinary mean is 14.5, dragged up by the 100. A 10 percent trim on ten values removes one from each end, the 1 and the 100, leaving 2 through 9, whose mean is 5.5, a far better reflection of where most of the data sits.
Related calculators
- Mean Median Mode: averages.
- Five Number Summary: quartiles.
- Standard Deviation: spread.
- Moving Average: smoothing.