Do you recall Beer and Nappies? … I never get tired of this legend, “diapers and beer”. Wal-Mart discovered through data mining that the sales of diapers and beer were correlated on Friday nights. It determined that the correlation was based on working men who had been asked to pick up diapers on their way home from work. wow. this is just a drop in the ocean of success stories about data. it is no more a secret that those who are able to objectively look at the data indeed see beyond sight. let me put this simply, there is nothing that leaves the defendant speechless as when the accuser says “Data does not lie.“. this put the last nail on an argument that was going nowhere. but then you must correctly refute this claim by asking “if data does not lie, what on earth is going on with all those election polls?“. we’ve seen the polls getting it wrong in the US in 2016, we’ve seen this recently in Australia. and now in Europe. this obviously couldn’t be an outlier event? could it be? no one knows and we may not know. nevertheless, I will give my best sample space of hypothesises.
My first Hypothesis is, maybe with the rise of social media and the fact that the news industry is struggling to compete with nowadays self-made social media journalists, analyst journalists tend to hurry to bring the breaking news. no time to formulate their polls approaches and methods and hence they end up publishing results that are nothing other than gut feeling and best guesses.
My second Hypothesis aligned with the first one is that maybe polls have been traditionally used by the media to influence people’s opinions. but with the social media now leading the dance, and people interconnected that never before, the trick is no longer working.
My third hypothesis is that politicians have understood that the data does not lie indeed. and so knowing in advance that they are facing a brutal defeat, they do everything they can, they spent millions on data analytics farms to find the one thing that will stick in people’s heads and hence tip the balance of votes into their favour. imagine how fearmongering politics is efficient nowadays. Imagine having the data that hints about the worst fear in people’s mind? sounds to me the best way to play the data against itself.
Regardless of how many times the data will get it wrong, the truth remains the truth. although a lie could prevail for some time, always trust your data and always ask questions. remember, the data is shy, it never talks unless you ask. but when you ask, you will be surprised to find out how talkative it is.