I found this interesting infographic regarding 2009 United States bankruptcy and foreclosure data at Americas Gone Bankrupt via Bookmark Share.

click image to expand

I love the description placed at Bookmark Share: “2009 United States bankruptcy and foreclosure data both visualized and broken down into depressing facts.”

Depression aside, there are a lot of effective elements in this infographic:

• The data are interesting.
• The decorative elements of this infographic do not obscure or distract from the data. All ornamental elements are located outside the main data display areas.
• The data are well labeled and the source of the data is included.
• I like the way the data are put in context explaining the frequency in seconds of a bankruptcy or foreclosure, how many occur per hour, and how many per day. If the point of the infographic is, “Wow, that seems like a lot per day.” The point is made very well.

What made me want to post about this infographic was its use of the map to display the bankruptcy and foreclosure amounts. Was this the best display technique?

Maps provide the unique ability to display information visually in relationship to its physical (spatial) location. Tables and graphs provide information on the quantitative aspect of data. For instance, for quick reference and precision, a table is usually the best choice. To make rapid comparisons of individual values or to look at the overall quantitative nature of the data, graphs provide a good option. To see how the data is distributed geographically, there is no substitute for a map.

Source: Robert Harris, Information Graphics: A Comprehensive Illustrated Reference
(New York: Oxford University Press, 1999), p. 226.

Let’s take a look at alternate ways to display the foreclosure data. Here is the data reformatted into a table and sorted largest to smallest. The advantage of this is that you now have information regarding which of the states have the most and least foreclosures. This is not as easily seen in the map.

Map Data Reformatted into Table

While there are too many entries in the graphic below, and I would probably try to reduce the number of rows depending on what question I was trying to answer, this better depicts the extent that Florida and California are outliers. Natural groupings also seem to appear in the data. You still have no sense of regionality though.

Click Image to Expand

While this map below was assembled quickly, is missing Alaska and Hawaii and some of the data labels are not shown, it is now easier to see some of the patterns in this data. It is clear that California has the most foreclosures followed by Florida. You can see a pattern along both coasts and up near the Great Lakes.

By showing relative size of foreclosures, you are able to take advantage of the spatial aspects of the data.

After seeing this data, I have all kinds of questions. Does foreclosure vary by state population? You could adjust the numbers so that you are looking at foreclosures per capita or perhaps add a data layer containing population.

Is there a correlation between foreclosures and population?

This map needs some clean-up, but there does appear to be a correlation between population and number of foreclosures. Looking at the numbers of foreclosures by population might be interesting. Perhaps it would be worthwhile testing for the correlation statistically.

Maybe it has something to do with how many houses there are in various areas?

Are foreclosures impacted by number of housing units?

To my eyes, this looks quite similar to the population overlay. Number of housing units may correlate with population – seems obvious.

Conclusion

While I think the original infographic is quite good, I believe the map with numbers provides less insight than either a table or bar graph, however, if you add in some graphical element that allows you to see patterns within the geography, the map can be far superior than the table or bar graphs.

What are your thoughts?

Comments