Thursday, July 10, 2014

From Shaun Gallagher, author of "Experimenting With Babies: 50 Amazing Science Projects You Can Perform on Your Kid" and "Correlated: Surprising Connections Between Seemingly Unrelated Things."  by way of Reddit.

Tell me ...
  • your sex
  • whether you have any children
  • whether you think pot should be legalized
  • whether you think you're a good singer, and
  • whether you like mayonnaise
... and I can predict a whole lot about you.
Your sex, Correlated's statistics suggest, can be used to predict your level of risk aversion (men, for instance, are more likely to want to skydive) and your capacity for empathy (women, for instance, are more likely to think we're too stingy with foreign aid).

Whether you have children tends to be a pretty good indicator when it comes to pop culture preferences (parents, for instance, like in-vogue artists such as Lady Gaga less than non-parents do) and, as you might imagine, youth issues (parents, for instance, approve of single-sex schools at a higher rate than non-parents).

Your opinion about marijuana legalization tends to be a good indicator of your political leanings and your opinion about a whole host of other social issues, from the death penalty to legalized prostitution to gay marriage, as well as your beliefs about religion and morality.

On average, people who describe themselves as good singers also tend to say they're good at a lot of other things, such as dancing or math or hula-hooping. On the flip side, people who say they're bad singers also tend to think they're bad at a lot of other things. So basically, this question tells us something about people's self-image and their perception of their abilities.

Similarly, people who like mayonnaise tend to also like a bunch of other foods, and people who dislike mayo tend to dislike a bunch of other foods. So this question tells us something about how picky a person is.

Given a person's answers to these five questions, I'd bet you could do a pretty decent job of guessing their responses to most of the other daily poll questions that have been asked on Correlated.
Still, day in and day out, Correlated is able to discover correlations that continue to surprise me — and I hope that they continue to entertain you.

No comments: