October 10, 2025

What Is A Trimmed Mean

Q: What is the difference between a trimmed mean and a median?

While both reduce outlier influence, a median is the middle value of a sorted data set (a 50% trimmed mean that discards all but the middle one or two values), whereas a trimmed mean removes a *specified percentage* from both ends and then averages the rest. The median removes extreme values implicitly, while the trimmed mean explicitly truncates a portion.

Q: When should I use a trimmed mean instead of an arithmetic mean?

You should consider using a trimmed mean when your data set is suspected to contain outliers or extreme values that you believe are not representative of the underlying phenomenon and would unduly influence the simple arithmetic mean.

Q: Does the percentage trimmed always have to be symmetrical?

Yes, conventionally, the trimmed mean involves removing an equal percentage of data points from both the lower and upper tails of the distribution. This ensures that the measure of central tendency remains unbiased.

Q: What is the relationship between a trimmed mean and a winsorized mean?

Both are robust statistics. A trimmed mean *removes* outliers before averaging, while a winsorized mean *replaces* outliers with the nearest remaining values before calculating the mean, thus retaining all data points but limiting the influence of extremes.

Discover what a trimmed mean is, how it's calculated, and why it's a robust measure of central tendency in data analysis, offering resilience against outliers.

Have More Questions →

Definition of a Trimmed Mean

A trimmed mean, also known as a truncated mean, is a statistical measure of central tendency calculated by removing a certain percentage of the highest and lowest values from a data set before computing the ordinary arithmetic mean. This process helps to mitigate the influence of outliers or extreme values that can significantly skew the simple mean.

How to Calculate a Trimmed Mean

To calculate a trimmed mean, first, sort the data set in ascending order. Next, determine the trim percentage (e.g., 5%, 10%). This percentage is applied to both ends of the data set: the lowest values and the highest values are removed. For instance, a 10% trimmed mean removes the bottom 10% and the top 10% of data points. Finally, calculate the arithmetic mean of the remaining data points.

A Practical Example

Consider a data set: [1, 2, 3, 4, 5, 100]. The ordinary mean is (1+2+3+4+5+100)/6 = 19.17. If we calculate a 16.7% trimmed mean (removing one value from each end), we first sort the data (already sorted). We remove '1' and '100'. The remaining data is [2, 3, 4, 5]. The trimmed mean is (2+3+4+5)/4 = 3.5. This significantly reduces the impact of the outlier '100'.

Importance and Applications

The trimmed mean is particularly useful in fields where data might be prone to measurement errors or extreme observations, such as economic indicators, survey analysis, or sports scoring (e.g., diving, gymnastics, figure skating where judges' highest and lowest scores are often dropped). It provides a more robust and representative measure of the 'typical' value compared to the simple arithmetic mean when outliers are present.

Frequently Asked Questions

What is the difference between a trimmed mean and a median?

When should I use a trimmed mean instead of an arithmetic mean?

Does the percentage trimmed always have to be symmetrical?

What is the relationship between a trimmed mean and a winsorized mean?