Read data

There is a saying : "Lies, damned lies and statistics". It told us how easy to read the number and how difficult to really derive something useful from it. Just want to share some simple tips about this

1. Have a right context

  • For example, you might read : "We sold this product to 10% of our stores" and think "crap !"

  • However, we might have several kind of stores (e.g. A, B, C) and that particular product was just designed for store A. The overall performance is 10% but we sold it to 70% of store A. Not that bad

2. Check data source

  • For example, you might read : "He traveled to 100/125 countries" and think "so cool !"

  • However, if we look further, there are just 25 stamps on his passport. And for the other 75 countries, he just "told" you. Then you need to dig further

3. Check sample size

  • For example, you might read : "Thong is the most handsome guy in the world, 75% people agreed with it"

  • However, turned out that I just asked 4 people for opinions, in which 3 are : mom, dad and my bro !

Notes: Those are simple tips that will help you to

- evaluate things (not just number)

  • and in some cases, to convince (read: mislead) people (e.g. my handsome example !)


Subscribe to Think.Forget.Do

Sign up now to get access to the library of members-only issues.
Jamie Larson