In the spirit of Freakonomics, Everybody Lies: Big Data, New Data, and What the Internet Can Tell Us About Who We Really Are tackles big questions with big data. For instance, few Americans would admit to being racist, but areas of the country with the most racist Google searches also voted for Obama less (adjusting for other factors). Google searches also suggest that the percentage of gay men is fairly uniform in every state, although surveys don’t agree (and who wants to out himself on a survey). And even if domestic abuse claims did not rise much during the last recession (somewhat of a surprise), online searches about abuse did, sadly.
Beyond the titillating examples, the author is careful to state the limits of big data, and in particular the fallacy of the random multi-variable analysis (thank you).