3 ms·
Six Degrees of Swine Flu
- dreish 18y agoSunday, the swine flu story breaks. On the radio the next morning, it's even bigger. I go in to work, and one person is home sick for the day -- someone I've never known to take a sick day. The next morning he still sounds awful over the intercom, and goes home sick again later in the day. True story ends. Now the hypothetical: Tuesday evening, I get a robo-call asking me to press 1 if I know someone with swine flu, and 2 if I don't. I pause for a minute; do I? I didn't ask him directly, though I joked about it in the usual detached office banter way, and I wouldn't expect he'd have got it diagnosed anyway. It probably wasn't serious enough to go to a hospital. But surely it had to be swine flu, right? So I press 1. I wouldn't do that, but I'm sure many people would.
- JacobAldridge 18y agoI don't see that this study or analysis takes into account that a person with the swine flu also knows 290 people. So all 290 of them would answer yes. Multiplying out 290 x the 290 people those people know to give some impressive number ignores the possibility you might actually need to be dividing. My math and probability analysis may be dodgy, but I hope my point is made: simply knowing how many people know someone in a category doesn't give any meaningful data about how many people are in the category.