Mean is sum divided by sample, whereas median is a line drawn in the sand with half above and half below. For normal distributions, they are the same, but for skewed data, like, ooh, say earnings, there's a fair chance they won't be. Odds on, median is lower than mean, as mean is biased upwards by high earnings, and no-one earns negative sums to balance things out.
Actually, that's not strictly true. Some people earn negative sums, like fat oaf Gordan Brown, they just aren't paid them.
no subject
Mean is sum divided by sample, whereas median is a line drawn in the sand with half above and half below. For normal distributions, they are the same, but for skewed data, like, ooh, say earnings, there's a fair chance they won't be. Odds on, median is lower than mean, as mean is biased upwards by high earnings, and no-one earns negative sums to balance things out.
Actually, that's not strictly true. Some people earn negative sums, like fat oaf Gordan Brown, they just aren't paid them.