Wednesday, July 23, 2008

Gapminder for All

I don't know if I heard about this and forgot, or missed it entirely, but the fantastic Gapminder software is now available for anyone to use as part of Google Spreadsheet (which is part of Google Docs). It is now a Gadget called Motion chart.

Here's an example. This is a portion of the data from my dissertation. It consists of emissions of unregulated chemicals by corporations, their revenues, and their lobbying and political contribution expenditures (adjusted for inflation).



I recommend setting the scales to Log, the y-axis to Lobbying, the x-axis to Weighted Emissions (which is the toxicity-weighted sum of a firm's emissions), the size to Real Revenue, and the color to Industry. There is an awful lot of noise in the data. Nonetheless, I think there is evidence of a weak pattern of northeast-southwest movement. That is, it seems to be the case that when firms emit more, they lobby more, and when they emit less, the lobby less. It is not possible to control for firm size rigorously in this graph, but I did include revenue as a variable in the regressions in my dissertation, and found that the tradeoff still persists. Granted, the explanatory power of the regression was fairly low.

But that's not really the point of this post. The point is to say "hey, here's a neat tool for visualizing data, and you can use it too!" A warning: I found the gadget to be a bit finicky. You have to be careful how you enter your time variable, for example (1, 2, 3, 4 won't do, but 1997, 1998, 1999, 2000 work fine). I hope that researchers and teachers will start using this.