December 18, 2012 - FILED UNDER Visualization
Demographics 101: Visualizations from the US Census Bureau
Several weeks ago, I received a questionnaire in the mail from the US Census Bureau that requested my personal and household information as part of something called the American Community Survey. I was unfamiliar with the project, but a bit of research revealed that the ACS is an effort by the Census Bureau to collect data in the years between the 10-year cycle of the regular census.
The depth and quantity of the questions on the survey got me thinking—what does the Census Bureau do with all of the information they gather? As it turns out, quite a bit. Last summer, the agency launched its Data Visualization Gallery, where a new visualization is posted every week, using historical and current census data.
In addition to producing their own graphics, the Census Bureau also makes much of their data available to third party designers and researchers via an application programming interface, or API. These outreach efforts combined with effective inhouse visualizations have served to expand the public’s perception of the agency and reshape its role and mission—a lesson applicable to all organizations and businesses.
Let’s take a look at some of the visualizations produced by the Census Bureau (and those making use of their data) and explore how population and demographic statistics have been transformed into engaging and informative graphics:
The first offering from the data visualization gallery is simple enough—a word cloud made up of US city names, arranged geographically. At first glance, the size of a particular city’s name appears to reflect its total population. But a closer look reveals a more interesting conceptualization: name size actually depicts the number of times the city has been listed as one of the top twenty most populous cities (nationally) in any census since 1790. Thus a city like San Francisco whose population exploded during the California Gold Rush in the mid-19th century is featured more prominently than Los Angeles, which was sparsely populated until a migration boom beginning in the 1920s.
Clicking on a city allows you to pull up a separate graphic charting its appearances on that top twenty list and changes in rank over time. Each of these graphics represents an intriguing narrative about economic growth and population migration. Detroit (pictured above) was one of America’s seven largest cities for nearly eight decades as a result of the thriving automobile industry and the influx of new workers from Europe and the American South. But the chart also depicts the decline of the city’s auto industry since the 1990s, as well as the flight of its existing population to the suburbs.
What appears to be a simple word cloud visualization is actually an interesting take on population change, exhibiting historical change tacitly in a new and effective way.
The gallery offers a series of visualizations called population density profiles. These graphics let users explore population figures within a five-mile radius of major US interstates, such as I-5 along the West coast and I-95 along the Eastern seaboard. By dragging a switch along the side of each map, the user “drives” a car from south to north; population data appears on a separate graph along the side of the map as the “car” moves up the highway.
These clever graphics offer two simple visualization lessons: First, the importance of a dynamic presentation. Although the data depicted in these maps could just as easily be shown in a static fashion, the interactive elements of the graphic and the animated depiction of travelling really helps to convey how these road networks move in and out of dense pockets of population. Second, there is something to be said for breaking the interstates down into separate visualizations, but keeping them similar enough that they can easily be compared. Although both I-5 and I-95 follow the coast from north to south, there are profound differences in population density that can been seen in exploring each map separately.
Perhaps the most interesting use of census data comes in the form of a New York Times project—using the Census Bureau’s API—titled “Mapping America: Every City, Every Block.” Using data from the American Community Survey collected between 2005 and 2009, the New York Times has created a fascinating, interactive graphic that does exactly what the project title claims.
Users can type in a zip code for anywhere in the US, and explore a map of the area based on a number of demographic factors, including race, income, and education level. In addition to all of the factors built into the map, the extent to which users can change their view of the map allows for additional exploration. Population density, for example, really stands out on this map—just compare Manhattan and Boston and you’ll see what I mean.
All of the Census Bureau’s graphics are posted on Facebook and Twitter, where they can easily be disseminated and discussed. This serves to remind the public (and Congress!) of the important work the bureau performs even in the years between its official decennial population surveys. Indeed, the agency describes itself on its Twitter account as “your trusted source for quality statistics about our people, places, and our economy”—a role far broader than head counter.
Effective visualizations using Census Bureau data are thus part of a larger branding effort to change the public’s perception of the agency, and to make American Community Surveys feel less like a burden and more like an opportunity to participate in a historical project. The agency’s proactive efforts in this transformation are commendable, and offer some clear lessons to businesses large and small looking to expand their own image and mission.
Image Source: www.istockphoto.com