Email Doesn't Bring Us Any Closer Together
from the still-six-degrees-of-separation dept
Apparently, email doesn’t bring the world any closer together. Some researchers decided to repeat the famous “six degrees of separation” study, where people were asked to try to contact someone else at random, by going through people they knew. On average, it turned out, that there were six degrees of separation between any two random people. The new experiment tried to see if that result changes as a result of the internet and email. The results showed that, even with email there’s still an average of six degrees of separation. The study also discovered some other interesting findings about what sorts of people were best to contact. Very few people used “internet only” friends (which probably explains why there’s no difference when done over email). Starting with people you knew at work seemed to yield better results. Also (oddly) when contacting members of the same sex, there were higher success rates. The most interesting point, though, was that most people didn’t rely on “social hubs”. There’s been plenty of research about the importance of social hubs in social networks, but this research suggests they might not matter as much as people think. If you think about it, this makes sense. If you’re asked to contact someone specifically, you’re always going to think of the specific individual who might know that type of person best. Falling back on the “social hub” that you know is sort of an “I give up” position. It means you can’t think of someone specialized enough to perhaps know the person in question, so you might as well try the shotgun approach and hope the social hub has a better idea of the target person.
Comments on “Email Doesn't Bring Us Any Closer Together”
couldn't they have tried that on friendster?
or tribe.net, or any one of those? or did they specifically avoid non-email methods. (i’m studying the effect of the internet on social structures right now, and this question turned out to be very important)
Re: couldn't they have tried that on friendster?
Considering Friendster is just a few months old, and Tribe.net is a few weeks old, I imagine this study was done before either existed. However, I’m not sure the results would have been that different.
What about six degrees of hate?
It would be interesting to see if the internet has spawned more hatred than friendship. In the past, I’ve received death threats from a Norwegian who didn’t like my opinions of Buddhism, so he called up my house and stuff. There are the Arab men who indiscriminately approach men and women for sex, then get all offended when they are rejected. Another Azerbaijani girl I once talked to swore that “Americans invented sex”, so I made fun of her, asking if Azerbaijanis used to reproduce by dividing in two, like bacteria. You can guess how she feels about me now. I asked a Serbian girl if any NATO bombs fell on her house. Another Norwegian nurse vacationing in Spain was gloating about her vacation, so I asked her how she felt about people in wheelchairs who can’t go outside. She got very angry and said people in wheelchairs should be killed.
Six degrees of Sep - VERY BAD STUDY
The study you report on is obviously terrible and should not be taken seriously. Determining how closely people are actually linked, or separated, requires enumerating all of the contacts of millions of people. Monte Carlo methods (which is what this study really used) do not emulate reality here.
In particular, people may be less likely to know whom their email contacts are in email contact with, than whom their friends are friends with, so how can they guess efficiently how to get an email from person X to person Y?
A better experiment would be to monitor a few billions of pieces of email, make lists of senders and recipients, and calculate how the minimum degrees that separate every possible pair of them.
– The Precision Blogger