Everybody Lies by Seth Stephens-Davidowitz

Whenever I visit a new doctor, I am always handed a questionnaire while I sit in the reception area. This form inevitably asks two questions: how often do I exercise each week and how many servings of vegetables do I eat each week? And to be honest, my answers are probably a little untrue.

This form is for my doctor and no one else. It helps them provide me with the best care possible. Then why do I fib on this private form?

Because we love to lie to ourselves. We all think we are better people than we actually are. We all think we are better drivers, better looking, and smarter than our average peer.

Everybody Lies is about finding us in the real data. So where are we the most truthful? Apparently, the internet and especially on Google.

There is a lot of interesting information here. Can internet searches really predict a presidential election? Can they predict divorces? Can the words in a loan application determine who will most likely pay it back? Possibly.

The book is very interesting, however, it completely collapses on itself in the last section. The author shows us that with a large amount of data it is easy to find a misleading correlation. If you flip a thousand coins every day for 2 years, one coin will show a strong correlation with the economy. Obviously, there is no causation between the two, but you can see how easy it is to be misled.

So in the end, this book is a great introduction to the power of big data. We should definitely use it but with great caution. This was a quick read with funny bits and interesting information.

And yes, I finished the book.