@realjohnboy,
sure. There are infinite ways to do it. It just means that the median is closer to the lower values and there are some up-jumpers that pull up the mean. With n=5 the median will always be the 3rd value when the values are sorted lowest to highest. With n=12,000,001 it will be the 6,000,000 value. You could have 5,999,999 folks who range from 1 - 22 and 4,000,000 who range from 22-44 (symmetrical) but the other 1.9 million that have higher numbers will pull up the average without affecting the median. In this case it just turns out that the average was pulled up to 40. You'll oftentimes see this with a bi-modal (two bumps) distribution where the median sits in the lower higher bell but there's another (smaller) bell sitting further down the x-axis on a graph. This second bump will affect the mean but not the median. Here's a random graph of a bimodal distribution. The median here is probably 123 or 124 but the mean would be up in the 140s.
Skewed data will have the same effect on the mean and median with the asymmetrical tail on the upper end pulling up the mean.
Here's a graph that shows the mode (most common value), the median (center value) and the average (x-bar) of a right skewed set of data.