Demo: Estimate Correlations between Smoking Rate, Cigarette Tax and Beyond

From Data-gov Wiki

Jump to: navigation, search

Infobox (stable demo) edit with form
  • name: Smoking Rates and Taxes Mashup

live demo here


Facts about this Demonstration

Live Demo(s)
Video Demo(s) Data source(s)
Other Data source(s)
Technology Used
Related SPARQL
Related Demo(s)

How to use

Pick two variables of the six provided to compare. You can do this by:

  • Using the pull-out lists on the X or Y axis of the scatterplot to the left.
  • Clicking on the appropriate cell in the correlation matrix.

  • On the left is a scatterplot of your chosen variables. The quantity one variable is the Y axis, the quantity for the other variable is the X axis. Each point represents a state.
  • On the top-right are two maps, one per variable. The darker a state is, the higher the value.
  • In the middle is a gauge showing the correlation between the two variables (1 - strong positive correlation; -1 - strong negative correlation; 0 - no correlation)
  • On the bottom-right is a table showing the correlation between every pair of variables.
Please note that this demo currently does not work properly in Internet Explorer.


Cigarette tax and the ratio of smokers to non-smokers have virtually no correlation (0.06). Rhode Island has a high smoker/non-smoker ratio, though its cigarette tax is also high. Maine has the highest smoker/non-smoker ratio, and its cigarette tax is in the middle. New Jersey has a high tax and a relatively low smoker ratio.

Unsurprisingly, population is strongly correlated with the number of smokers, estimated total number of smokers based on survey data, and the number of correspondents. It has almost nothing to do with the smoker/non-smoker ratio or cigarette tax. In fact, those last two variables don't correlate with any of the others.

Technology Highlights

It creates a Google datatable via several SPARQL queries. Some variables are created by mashing up earlier variables (e.g. smoking ratio is (Smokers / (Correspondants - Smokers)). The final table is then passed on to the API for processing and visualizing.

Personal tools
internal pages