Everybody Lies

The name of the book drew me in and I made an effort to read Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are, a book on the aforementioned topics presumably for the lay person. Sadly I lasted only until the middle of chapter one - I did not think I would learn very much from sticking with it to the end. 

The author, Seth Stephens-Davidowitz begins by stating that big data analysis of note is usually explainable to a reasonable person. I would totally agree with that when it is not, we are probably dealing with a peddler of snake-oil. He went to explain that grandma offering relationship advice is indeed big data. She has the most data points on the topic and has observed outcomes over the long years. Based on that she can see patterns and offer a recommendation for the person at hand. A very reasonable explanation of what data and analysis is all about.  

Somewhere in chapter one the author proceeds to make a liar out of grandma based on analysis of relationship data from Facebook (God help us). The big reveal is that when two people are part of the same social circle their relationships tend to end. From that one may conclude, having separate and apart friend groups make for stronger, longer lasting relationships. Grandma had never figured that one out. 

Is it possible that in the fishbowl of a common friend circle even a strong relationship is put through a severe stress test and maybe that contributes to its eventual demise. Whereas when the friend groups do not overlap there is more room to breathe. Was there a test group that was not on Facebook at all and how did they fare with having common friends or not ? What about people who do not update or correctly reflect their relationship status? Its easy to throw poor Grandma under the bus but I think her recommendations are way better.