Thursday, 18 August 2011

Yahoo Research - Facebook six degrees / small world experiment

In 1967 Milgram and Traver's undertook their Small World Experiments.  They measured how many steps it took for a package to make it's way from a randomly selected individual to a target elsewhere in the US.  On average it took six steps for a chain to complete - a concept that was subsequently embodied in the idea of everyone on the planet being 'six degrees of seperation' away from each other.

However, the sample size was small and the high fail rate is often overlooked.  Is everyone really linked by six steps?

In 2008 Jure Leskovec and Eric Horvitz studied Instant Messenger data to see if they could find the same effects, analysing 30 billion conversations from 240 million different people over the course of a month.  They found that again the average shortest path length was around 6 (median 6, average 6.6), though sometimes they found longer paths, with the longest showing 29 steps (this research can be found here.)

Yahoo! Research small world experiment
Now it's the turn of Yahoo! Research.  They are taking the best of both of these, setting the spirit of the Milgram experiment against digital infrastructure, but this time they are using Facebook.  The new Yahoo! Small World experiment offers participants a target and then asks for them to choose one Facebook friend to help move the communication to that target.  There is also the opportunity for users to nominate themselves as a target.  Will the results show that Facebook has made the world a smaller or larger place?

Interested to see what they find!

Join in at: http://smallworld.sandbox.yahoo.com/

Related posts

Six degrees of seperation - Stanley Milgram, social networks and experiments on LinkedIn

Like this post?  
Then subscribe to regular updates from this blog - 
click here to use a Reader or click here to get email updates

3 comments:

Jake said...

No matter how you look at it, there is no denying that the world is becoming much, much smaller!

Mat Morrison said...

@jake - I suspect that there are ways one could look at it where the world *isn't* becoming smaller;

While it's true that information (news, trends, fashions, urban legend, rumour) travels faster, it's not necessarily the case that people are any more connected than they were at the time of Milgram's experiments.

So this leaves us with an interesting problem; the world *seems* much smaller, while our personal networks and spheres of influence remain very similar. Does this lead to a reduced focus on what may practicably be achieved?

Nick Burcher said...

@jake @mat Thanks for comments, interesting debate starting. Is the world smaller than it was?

I think probably - if nothing else it's more searchable.

Does this mean that it is easier to spread things?

Yes and no. In theory things can spread much further, but in reality the pool of things that can spread is much larger (and information / content is not perishable now, it is archived online so even more noise to cut through.) So greater connectivity, but greater choice at the same time. There is also the problem that receiving something does not necessarily mean that it will be sent on.

So the hits get bigger, but arguably they are also more unpredictable. For every Susan Boyle there are hundreds of Joe whatever his name was - just because it can doesn't mean it will and I think there is a gulf between what can be achieved in theory and what can be achieved practicably....

Interested to see the Yahoo! results.