Data Designer

Blog

Facebook vs. Princeton

Last week, researchers from Princeton released the results of a social network study, detailing how models based on Google search query data revealed that Facebook would lose 80% of its users by 2015-2017.

We applied the model to the Google data for search query Facebook. Extrapolating the best fit model into the future suggests that Facebook will undergo a rapid decline in the coming years, losing 80% of its peak user base between 2015 and 2017.

This is a bold statement indeed, and a little hard to swallow. Facebook wasn't buying it, and questioned Princeton's use of Google search data rather than actual engagement trends to draw their conclusions, and so did a little research themselves. By applying the scientific principle seemingly adopted by Princeton, the classic correlation equals causation, they found some startling trends of their own.

In a direct response titled "Debunking Princeton", Facebook published the results of their own study, in which they found a strong correlation between between undergraduate enrollment and its Google Trends Index. 

Unfortunately, this is not all good news. Princeton’s Google Trends search score has in fact been in decline over the past few years, extrapolating this out a few years we can see that by 2021 there’ll likely be no students left at Princeton at all! Facebook then goes on to state their concern for the planet, as Google Trends for "air" has also been declining. Extrapolating that to 2060, we can conclude that there’ll be no air left on our humble little planet. Yikes!

Facebook's point is well-illustrated, correlation simply does not imply causation, and their debunking article is a great reminder on how easy it is to make this mistake.

As data scientists, we wanted to give a fun reminder that not all research is created equal – and some methods of analysis lead to pretty crazy conclusions.