The Pitfalls of "Big Data"
I greatly enjoyed this article ("Big data: are we making a big mistake?", by Tim Harford) on the dangers of "big data", which certainly has opened up new avenues of research but nevertheless still requires understanding to avoid statistical pitfalls. Understanding causes beyond mere correlation is still necessary: correlations can predict trends with uncanny accuracy, until they break down.
Very useful is this idea of "digital exhaust" : "…we might call 'found data', the digital exhaust of web searches, credit card payments and mobiles pinging the nearest phone mast". I rather like the phrase, too.
Read the article and heed the warning: "But while big data promise much to scientists, entrepreneurs and governments, they are doomed to disappoint us if we ignore some very familiar statistical lessons."
In: All, It's Only Rocket Science, Plus Ca Change...