Wisdom of the crowd vs Regression to the mean

I am a big fan of Last.fm. Over the past couple of years it has become my default way of finding out about music. The way it learns your taste and (somewhat) challenges it is great and I have enough of the trainspotter about me to be interested in the charts it creates of the music I have most listened to in the past week, month or year.

Last.fm, like everything in this world these days, has a social networking component, part of which allows you to join groups. I belong to two such groups, one for readers of Metafilter, the other for people who shop at PostEverything.

So, two groups, one aimed at a mainly American technology aware, socially minded group of people; the other for people who shop at an esoteric online shop for “interesting” music.

And when you look at the charts of most popular music in the two groups you find … a large degree of similarity. Radiohead … check, Coldplay … check, Portishead … check.

There is nothing surprising in popular music being popular. But it got me thinking about taste and popularity and the connection between chart position and tag clouds. One goes up and one gets bigger but both tell the same story.

And the concept of popularity always makes me think about the story of how architects define paths on campuses by letting people walk across the grass, organically creating links and these emergent paths are then used as the basis for the formal paths.

This is often quoted as an example of the “wisdom of the crowd“. The way in which a decision taken by a group may be smarter than that taken by individual members of the group.

But you can also think of it as an example of “regression to the mean“. The tendency of distributions to collect around the mean point. This is just a natural reflection of the fact that most things are around average, by definition, but we humans like stories, we like to think that what we are seeing is something real, not just a statistical artifact.

We see this time and again in life. A football manager is acclaimed one season as a genius and accused of failure the next, when you look at the stats all that has happened is that they won slightly more games than the form book would indicate in year one and slightly less in year two. Whenever things are at an extreme we either proclaim a new paradigm – “The Bubble is eternal!” or start looking for reasons to explain something which may just be noise.

I find myself thinking about what this might mean for social media and engagement. If we create a more agile public space, one which is as frictionless as possible. Where people can rapidly engage with and shape policy, is there a risk that we conflate noise and reality?


One thought on “Wisdom of the crowd vs Regression to the mean

  1. You are absolutely right and it is something that really, really needs to be thought through. It will happen, people will expect it and if government doesn’t do it, someone else will. Policy officials and Ministers need to have a jolly good think about what they are going to do about this, when and how.

    The answer will lie somewhere in the compromise region, I am guessing. Policy can be put in a space where engagement and shaping might happen, but how much of an influence this would have on the final published policy would have to be played with. And at what point is this policy put ‘out there’?

    Am going to go away now and think. Mitch Sava is very good at this stuff

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s