22-25 April 2026

Average Fail - Why Averages Can Be a Terrible Idea in Data Analysis

Short 20 minute session for SQLBits 2026Saturday - 25 Apr 2026 - 14:40 - 15:00 Room 1B

TL; DR

Averages can bite you - repeatedly. Come see how and why, and have a laugh and a cry at the same time!

Session Details

What is the easiest way of getting a number controlled for various factors? An average.

Most people stop thinking about the consequences right about here, but the harsh reality is that in many cases, an average is a terrible idea - especially if you don't know the data you're working with in the first place!

Consider what would happen if you start using averages to fill in missing data. or the potential for disasters with trying to average qualitative data expressed as numbers.

I'll spend 20 minutes raking the concept of averages over the coals. Bring marshmallows.

3 things you'll get out of this session

Why averages are bad
How averages can skew results
What Chicago adds to the story