We briefly reviewed the measures of central tendency that you’ve known for years (mean, median, mode) and then talked at length about how to describe how “spread out” values are.
We noted that two distributions of values could have very different shapes (remember the tall-skinny vs. short-wide graphs, and the two-bump camel graphs) but have the same mean, median and mode. In short, those aren’t always helpful in describing our data sets.
What we want to be able to do is describe how spread out our data is. Another way to think about it: if you choose an individual at random from your population, how likely are they to be far from the mean? If you’re data is spread out, they’re more likely to be far from the mean.
This is what standard deviation is for: it lets us describe the spread in a numerical way.
If you were away, or if you want a nice review of these ideas, check out this link:
This is a nice, clear explanation of standard deviation including pictures of dogs. If you have a friend with a Dachshund, you might want to show her too.