When should you prefer the median over the mean as a measure of central tendency?

• A. When the data are normally distributed
• B. When the data are unimodal
• C. When the data are skewed or contain outliers
• D. When the sample size is large

Answer: (C)

The essential point is that the median is more robust to extreme values than the mean. The median is the middle value when data are ordered, so it doesn’t get dragged toward very large or very small observations. In skewed distributions or when outliers are present, those extreme values pull the mean toward the tail, making it a poorer reflection of a typical observation. The median, by staying near the center of the bulk of the data, better represents a typical value in such cases. For normally distributed data, the mean and median are alike, so there’s no strong reason to prefer one over the other, and large sample size doesn’t inherently require using the median. Unimodality describes shape, not the impact of outliers on central tendency. Therefore, you prefer the median when the data are skewed or contain outliers.