Marianela Luna-Torrado '24
The visualization above has one dot for every issue of The College News in which ‘negra’, ‘negress’, ‘negritude’, or ‘negro or negroes’ is mentioned. Each word has 3, 4, 2, and 440 dots, respectively. Click on the dot next to a word in the key to display its dots only. Hover over a dot to view a popup with the issue date, mentioned word, word count, and word context. Find the issue by searching for its date on The College News collection or in the corpus text files. To zoom, pinch in or out on a computer touchpad while hovering over the visualization.
With the resurgence of the Black Lives Matter movement last summer and the Bi-College strike against institutional racism the following fall, I wanted to learn more about how the Bryn Mawr community historically perceived and treated black people.
First, I used Voyant to look through the corpus for race-related words. The word ‘negro’ and its related forms appeared several times, hence my project focus. To study these words, I wrote and ran a script that searched every issue for a word and its count. I also wrote and ran another script that searched every issue for a word, and if found, identified its context by selecting 100 characters before and after. Eventually, these two scripts were combined into one script, which I edited so it did the functions described previously plus selected the first and last contexts for issues with more than one instance of a word. All of this information was saved to a CSV file.
Afterward, I cleaned the CSV file in OpenRefine. This involved fixing errors like duplicate contexts and formatting certain cells so they were more readable. Lastly, I used the Python library Altair to visualize the cleaned CSV file as a table bubble plot.