What is the primary advantage of using median imputation over mean imputation for a feature with a skewed distribution?

The mean is highly sensitive to outliers, which can significantly shift the imputed value in skewed data. The median represents the middle value of the dataset, making it a more robust measure of central tendency when outliers are present. While both methods reduce variance, the median specifically mitigates the distortion caused by extreme values, whereas the mean would be pulled toward the tail of the distribution.