Skip directly to content

Geography of Diabetes in US: A conditioned map

on Mon, 10/01/2012 - 01:29

This interactive data visualization titled "The Geography of Diabetes" allows to explore the magnitude and spatial distribution of Diabetes across US Counties. It was created by Annette Greiner using the core set of indicators from the Food Environment AtlasUS Department of Agriculture (USDA)

What I think it is very interesting in this visualization is the implementation of a Conditioned Choropleth Map. The Conditioned Choropleth Map, a technique originally proposed by Professor Daniel Carr,  can show the association between a dependent variable and two or more potential explanatory variables. The data-analytic objective is to promote better-directed hypothesis generation about the variation of a dependent variable. The Conditioned choropleth map approach does this by partitioning the data into subsets to control the variation in the dependent variable that is associated with two or more conditioning variables.

The above data visualization display a map of the Prevalence of Diabetes by counties in US which is the dependent variable. Other three indicators, Poverty rate, Adult obesity rate and Percentage of white population are used as independent variables. They are used to filter counties on the map based on their values. By default, all countries are displayed meanig that the complete ranges of poverty rate, adult obesity and % White population are included. So if using the sliders located at the bottom-right side of the visualization, the range of any of the three independent variables is changed, the number of displayed countries will vary accordingly. For instance, changing the range of poverty rate from 3.1 - 53.3 to 20 - 53.3 the map will display only the counties with poverty rate in this new range, note that most of the displayed counties have higher diabetes rates giving information about the positive association of poverty with diabetes.    

The Conditiened choroleth map is a simple and very useful technique for exploratory data analysis hypothesis generation. 

Enjoy it!!! 

Ramon

Post new comment