By Ross Perez 06 Avr, 2010

It is no secret that obesity is a huge problem in the US. Data on the subject is abundant and we used some from the CDC to create the following visualization. However, the next step in the fight against obesity will be transforming data like this into useful policy decisions. With that goal in mind, the Department of Health and Human Services is creating a massively updated version of the CDC database that will allow policymakers to see health indicators as here, but it will also include detailed info on procedures and costs by institution that will hopefully lead to more purposeful spending. We created this viz in anticipation of this upcoming data and to encourage as many people as possible to "viz it" with Tableau Public.

As you might expect, obesity is highly correlated with smoking, levels of physical activity and the amount of fruits and vegetables one eats. What is interesting here are the differences between regions. In general, the midwest has significantly weaker correlation for all three contributors. One must assume that there is a hidden variable (more physical activity that is not exercise?) keeping obesity rates lower there.

What happens when you open a rich data set, like the upcoming HHS release, to analysis by anyone? We don't know, but we are counting on institutions like Anil Dash's Expert Labs to create innovative solutions and "help policy makers in the U.S. Federal Government tap into the expertise of their fellow citizens." The open government movement has begun, now it must be supported.

What we like about this viz

Highlighting: Since every point is on top of or near 3 or 4 other points, there is significant overlap. Without highlighting, this viz is a large blob. With highlighting, this viz is a useful analytical tool.