Which measure of central tendency is least affected by outliers?

• A. Range
• B. Median
• C. Mean
• D. Mode

Answer: (B)

Outliers don’t change the middle value when data are lined up in order. The middle value, or the median, is determined by position rather than by how large or small the numbers are. That makes it robust to extreme values.

For example, with data 1, 2, 3, 4, 5, the median is 3. If you add a huge outlier like 1000, the set becomes 1, 2, 3, 4, 5, 1000. The median moves only to the average of the two middle values (3 and 4), which is 3.5, whereas the mean jumps from 3 to about 169.1. The mean is pulled toward the outlier because it incorporates every value. The mode and the range relate differently—mode is about the most frequent value, and range is about spread—so they don’t consistently reflect central tendency the way the median does in the presence of outliers.

So the median is the measure of central tendency least affected by outliers.