A YouTube for data

I admit it, I'm a data junkie and I love economics. Not just all that stuff you hear on CNBC or coming out of the political talk shows, but applying economic analysis to everyday problems.

That's one reason why I'm such a huge fan of the Freakonomic's book and blog.

Now there's a sharing site for those of us addicted to data. Swivel allows you to upload data, graph it and share it with other people in much the same way that YouTube or Flickr allows you to share video and pictures. You can tag data, rate it, compare it, post it on your blog and have it automatically updated, all kinds of useful things. The amount of content is a little light right now, but I am forecasting many late night in front of my computer looking at all the weird combinations of data as content gets out there. It's also a great way for me to find information from all types of sources to support hypothesis.

Of course a lot of people will begin attacking the validity of the data, and it should be looked at with some skepticism. However, it should not be ignored, if you can validate where the data is from and that it's accurate, this can be a powerful tool.

Here's one of my favorites so far. Comparing violent crime to wine consumption:

Drink more wine, keep violent crime down. (link via Freakonomics blog)

- Paul Herring

Comments

[Cam Beck]

This is a classic (and fun) example of a post hoc fallacy, as pointed to in Darrell Huff's "How to Lie with Statistics."

According to Huff, "There is a close relationship between the salaries of Presbyterian ministers in Massachusetts and the price of rum in Havana."

"Which is the cause and which is the effect? In other words, are the ministers benefiting from the rum trade or supporting it? All right. That's so far-fetched that it's ridiculous at a glance. But watch out for other applications of post hoc logic that differ from this one only in being more subtle."

Good book.

[Paul Herring]

You're right, however there is a 'related' factor that shows how variables are related based on the number of data points and the correlation.

You have to really look at the data, however, before you rush to judgement. What I like about this site is that it allows people to create, share, comment and rate data. My hope is that it becomes 'self-policing' sort of like Wikipedia.