Calculating statistics on whole numbers

Steve Simon


This page is currently being updated from the earlier version of my website. Sorry that it is not yet fully available.

Dear Professor Mean, I have some data that has a resolution of 1, meaning that these are whole numbers only and no decimals. When I calculate the mean and standard deviation, should I report the actual value or should I round it?

There are some commonly cited jokes about this. The first is that if the average family has 2.2 children, what does the 0.2 child look like? There’s also the joke about how the average person has one testicle and one ovary and less than 2 legs.

In spite of these jokes, it is standard operating procedure to carry an additional decimal place for this type of data. In most situations, carry two extra decimal places in the mean or standard deviation is overkill. Any intermediate calculations, of course, should be computed to full accuracy to avoid round-off errors.

You can find an earlier version of this page on my original website.