In a recent study, people who sweated when the stakes were low did the best when stakes were high.
IN “GATTACA,” THE DYSTOPIAN cult classic set in the “not too distant future,” parents genetically program their children before birth, coding them for desirable strengths and skills. For them, biometric data is destiny: A person’s genetic code, tracked through a massive database, determines their career, which, of course, affects everything.
Nearly 20 years after that movie’s release, we are closer than ever to using biometric data as part of the hiring process, specifically to solve one chronic problem: Employers are bad at predicting who will perform under pressure. Each year tens of thousands of new Wall Street hires undergo boot camps that cost up to $6,000 a person, yet finance has a suicide rate 1.5 times the national average and the second- highest voluntary turnover rate (14.2%, after the hospitality industry). And if an industry as well-funded as finance struggles with vetting applicants, what hope do smaller businesses have?
For those who think it’s mathematically odd that Donald Trump was sworn in this past week as the next president of the United States — even though he lost the popular vote to Democrat Hillary Clinton by nearly 3 million votes — I have some news: It could have been even more strange.
Instead of netting only 46.1 percent of the vote compared with Clinton’s 48.2 percent of the popular vote, Trump could have, by my calculations, pulled in a mere 22 percent of the popular vote and still won the election.
How is that possible? Thank our quirky electoral college system, as outlined in the U.S. Constitution, that assigns electoral votes to final election outcomes in individual states, not by a nationwide vote tally.
We often hear that the Internet is unpredictable, that it’s the “Wild West.” That would seem to be especially true of a social medium such as Twitter. After all, tweets are by definition instant and short-lived. But in a paper I and my co-authors just submitted to the Annals of Applied Statistics, we describe a model we have developed that predicts how popular a tweet is likely to be within just a few minutes of when the “root tweet” is posted.
And anyone who wants to can now try out our model by visiting www.twouija.com.