Nathan Yau’s summary of 2013 at flowingdata is a festive must-read. Along with fun graphics like trendy kids’ names, he picked up on some trendy stats’ names too.
The hype around big data should settle soon, but it’ll probably be at least an additional year before we get to call big data just data again.
I’m not convinced by that. The penny may already have dropped among the numerate, but it’s guys in suits that are driving this by paying for it, and like dot-com, they will be happy to carry on paying just in case there’s something to it, in which case they don’t want the other guys to get there first.
Data science versus statistics? That’s going to continue. The good news is that it affects almost no one from a practical sense, so I’ll leave that argument to others.
I chuckled to myself at this. Every day – literally – I get emails from really hard working self-promoters who are pushing this vacuous data-science-it’s-not-just-stats message. I don’t want to name them in case you go and read their nonsense. “Great new article: what’s the difference between data science and science data? Subscribe now to read the awesome answer, written by a leading expert, me.” I don’t see this changing any time soon. It’s a sign of a trendy subject; I should rename this Robert Grant’s $tat$ Blog. But consider if you will the talk of booming numbers of graduates. Dot-com hurt some investors, but the web is in a way stronger place because of the boom in skills. If that happens to $tat$ it will be no bad thing.
There is a serious side to this data science though. What they talk about is looking at the data and thinking, not just following rules blindly and doing dull hypothesis tests. Which sounds like the difference between good statisticians and bad ones to me, nothing new there. If you want to read about data science, save your money and buy a copy of Cox & Snell’s 1981 (!) tour of good, thoughtful practice, “Applied Statistics: Principles and Examples”.