Digital history tools and methods have shaped spatial analysis, data analysis, and visualization of that data in profound ways. Many discussions of digital history focus on data analysis versus visualization. However, the two are more complementary than divergent. I have found data analysis tools often generate interesting visuals, and the process of creating visualizations of data analysis often creates new insight into the data creating new pathways for further analysis.
Straightforward examples of this are charts and graphs. The term visualization can seem complicated until you realize that it is simply referring to rendering data in a visual format rather than text. One of the most ubiquitous types of visualization is the data chart. Most people inside and outside of academia have seen and often created charts for school, work, or personal projects. Charts have a unique ability to render complex data accessibly.
Currently, I am conducting a place history of a small north-central Phoenix, Arizona neighborhood. I will be using charts as a powerful visualization tool to demonstrate factors driving change over time in this neighborhood. The neighborhood I am analyzing was part of the post-war housing boom that expanded Phoenix northward rapidly in the decades following WWII. Multiple factors propelled the expansion of the city northward, however a prominent driver of this expansion was affluent Phoenicians seeking larger homes on larger lots during the post-war economic expansion. As such, class factors heavily into the history of north-central Phoenix neighborhoods. Given this, consider the following chart:
Although outside of the specific time period I am analyzing for my place history, this chart (based on census data from socialexplorer.com), visualizes median income changes in inflation adjusted 2013 dollars over the 1990 to 2013 time period. It focuses on the census tracts north of McDowell Avenue and south of Northern Avenue in Phoenix. Pre-war, Phoenix ended at Thomas Avenue (Census tract 1118). Known as the Willow neighborhood, census tract 1118 has historically been wealthy. Census tracts 1105, 1171, and 1088 north of town pre-war were largely farming areas or locations where cheap lots could be purchased for home building. Post war, intense housing expansion began in census tracts 1075 (Medlock Place), 1066 (San Juan Marshall and Rancho Solano), and 1062. Medlock Place and Rancho Solano were marketed to wealthy homebuyers looking for larger lots and bigger homes in the late 40s and early 50s. Census tract 1062 featured even larger properties built in the 60s. Over time, the wealthy leapt over census tracts 1105, 1171, and 1088 seeking larger and larger properties northward. The data visualized in the chart suggests the effects of this leap frogging pattern persisted into the present with 1075, 1066, and 1062 retaining their wealthy inhabitants and 1105, 1171, and 1088 remaining poorer. The effects of income inequality over the last twenty-five years can also be seen as the richer census tracts largely retain and grow their wealth whereas median income in the poorer census tracts stays flat or declines.
This chart is a straightforward example of how data visualization can make complex trends readily accessible. For my place history, I intend to use charts to visualize aspect of change in the neighborhood like the ones shown in this chart. A specific example I am working on is charting the quantity of lots for sale as expressed in newspaper ads in the Phoenix Republic each year from 1945 to 1985 to show the size and pace of the post-war housing boom. Another, more narrow example, is to similarly chart the number of ads over time for lots and homes available in San Juan – Marshall to visualize how quickly the neighborhood filled in.
3D visualizations are another example of digital visualizations both making complex data accessible and creating new insight into the data. Although some may be surprised that 3D modeling is used in digital history, many historians have utilized 3D modeling techniques to great effect. The St. Paul’s Cathedral Project hosted by North Carolina State University is an example. The project recreated St. Paul’s Cathedral from the early seventeenth century using 3D modeling. The recreation allows viewers to experience what worship and preaching at the cathedral was like in that time. Specially, the project recreated the physical environment in existence when parishioners heard John Donne’s sermon for Gunpowder Day on November 5th, 1622 in Paul’s Churchyard located in the center courtyard of the structure. The project used 3D modeling to recreate both the physical space and texture mapping to emulate the building surfaces. This data was then used to acoustically model the space so that a recording of Donne’s speech could be made that would sound like how parishioners heard it during the actual sermon centuries ago. Recreating an environment like this using 3D modeling tools is particularly useful when you consider that the structure in question no longer exists. Old St. Paul’s cathedral burned down centuries ago. It is only accessible to researchers and the public through this virtual modeling project.
I am using 3D modeling in my place history as well. Here is an example:
Based on an aerial photograph from 1949, this 3D scene looks west from the modern intersection of San Juan Avenue and Central Avenue in Phoenix. From this vantage point, The Setter Farm is in the foreground with the Otter Farm to the South and the San Miguel subdivision in the background to the far west. The San Miguel subdivision was the first part of the San Juan – Marshall neighborhood plated and built. From a modern perspective, this entire area is all homes now. It is natural to assume it has always been like that. As I built this model and analyzed it from various perspectives, I settled on the point-of-view shown as particularly insightful. From this vantage point, the agricultural orientation of the area is quite prominent. Viewed from the San Miguel side, the model view seems to suggest a suburban dominance. However, viewed from the Setter Farm side you see that in 1949 agriculture still dominated the area. By 1959 this was no longer the case as all the farms are gone.
This is an example of how data visualizations can create new insights that lead to new avenues for research. Given the insights gleaned from this model, I intend to build further models off of newer aerial photographs to see what these visualizations tell me about the changes the area experienced over time.
As these examples illustrate, digital analysis tools and digital visualizations work together in a complementary fashion. The tools allow for data creation and analysis that often yields insightful visualizations. The visualizations, however, can also lead to new perspectives and avenues for further research. This powerful combination is continually shaping the practice of history.
“Virtual St Paul’s Cathedral Project.” Virtual St Paul’s Cathedral Project. Accessed November 20, 2016. https://vpcp.chass.ncsu.edu/
“Median Household Income.” Social Explorer. Accessed November 20, 2016. http://census.socialexplorer.com/pop-flash/
