Data Visualization

A data visualization of immunization rates, and research into visualizing uncertainty in data sets.

 

Info: Special problems independent study in Fall 2017.

Advisors: Prof. Alex Endert and Prof. Lauren Wilcox

Built with d3.js


The Data

The data set used for this project came from the National Immunization Survey conducted by the CDC on an annual basis since 1995.  This data is available on the CDC website[1].  The visualization looks at immunization rates for 24 month olds in the fifty states for the following immunizations: Polio and DTap.  More immunizations will be added in the future.  Data for years after 2014 will also be added as the data becomes available.

I chose this data set because the amount of uncertainty, in this case standard error, in a data point varied across states and across years. 

Viewers/Users

Potential viewers would be people inquiring about immunization rates in the United States, such as parents of small children or people interested in immunization data. 

Potential Questions that could be answered with the data set:

  1. "What was the rate of immunization for vaccine 'x' in state 'y' in year 'z'?"
  2. "Has the rate of immunization for vaccine 'x' fluctuated in state 'y' over the years?"
  3. "Has the rate of immunization for vaccine 'x' increased or decreased in state 'y' over the years?"
  4. "Does state 'y1' have a higher immunization rate than state 'y2' in year 'z'?"

 

Demo Video of Visualization

 

Initial Sketches & Brainstorming

IMG_5407.JPG
IMG_5409.JPG
Preliminary sketches of how to visualize uncertainty in the data set.

Preliminary sketches of how to visualize uncertainty in the data set.

Design Decisions

I chose to represent the data in this manner rather than a heat map because I wanted the visualization to show the data over a range of years, where a heat map would only be able to show the data for a year at a time.  A set of twenty heat maps, one for each year, could be displayed concurrently, but this visualization gives each state the same spacial weight (the color space for California = the color space for Rhode Island). 

Future Steps

I would like to do research more into visualizing uncertainty, and test interpretation of various representations on viewers to see if a specific representation is the most clear. I would also like to add a drop down menu and include more vaccines.  Finally, I would like to add a feature where viewers can change the order of the list of states so it can be sorted by region.  That way viewers can explore the data set and see if patterns emerge when states are sorted by region. 

Sources

1. https://www.cdc.gov/vaccines/imz-managers/coverage/nis/child/index.html