Missing the trees for the weekend

If your data looks like this:

A line graph showing noisy data, with the only clear trend being a dip for two every seven data pointsThen, congratulations! You’ve discovered that arcane effect known as “the weekend”. Now go normalize your data, and find out if anything interesting is happening.

I see this anti-pattern happen a lot. What interesting things are we missing because we can’t see past the main effect?

Of course, there is a danger of over-normalizing, and underestimating the impact of the main effect. If you can only afford to staff the help desk five days out of seven, and you’re trying to figure out which days to close your business, then by all means ‘discover’ the weekend…


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s