Experiment Question: Most scholars have argued that the volunteers largely came from middle-class backgrounds. How might we use ArcGis Mapping to confirm and/or complicate this claim? 

The Social Demographic Map visualizes the home locations of (select) volunteers on poverty rates at the county level (1960 Census Data). The inspiration for these maps comes from the work of SNCC. Explore SNCC’s report using census data or research department to learn more about the ways fieldworkers used census data in their work and how they used the data to map the projects for the summer. This website ran an earlier experiment grouping income level data and population counts based on racial identity. See the report here  This early work, however, was flawed because the “nonwhite” category marker used by the 1960s census was misleading when grouped with income data. This current iteration focused on income level data. 

Methods and Data: This map draws on two sets of data: 1960 county-level census data on income levels and the home locations of the volunteers. I first joined the two data sets, intersecting the home location of the volunteers with the county data on income level (percentage). Using grouping analysis, a Ripley K Function method in ArcGis, I created a map that generated 6 “groupings” based on the income level of population counts at the county level. Tract level data would be ideal, but the available 1960 census data only provided relevant data at the county level. The current categorizations in the legend are my creation.

Results and Discussion: The website needs to revisit the construction of the data files on the income level to make sure it accounts for the size of the household. Thus, the website will run another experiment with Tract Level data (on race and income) and update home locations of volunteers. Once this data is available and the experiment is run, the website will be updated accordingly.

Future Question (s): Using the different economic categories of where volunteers came from, are there differences among students in how they interpreted the summer (or poverty in Mississippi) based on their economic backgrounds (county-level)? Alternatively, especially based on the above limitations, is there a different way to run this analysis? Both questions seek to raise a larger methodological question for digital historians: How might a macro analysis (e.g. census data on poverty ) aid historians in their qualitative analysis of the volunteer experience (e.g. diaries, letters, home, etc.)?