Knowledge-seekers’ interaction with archival records and primary source materials are often defined by barriers. Access to materials is by appointment only and direct browsing of materials is prohibited for fear of damage or unintentional migration of a box’s contents. Digital collections can provide a glimpse of an archives’ holdings, but usually represent a small portion of the entire collection. Even the existing guides to the collections present obstacles to access due to their inconsistency in tone, use of jargon, or tendency to favor archival structure over more user-friendly organization. Inspired by the existing visual story-telling work of teams from the California State University of Bakersfield, University of Illinois Champaign-Urbana, and Princeton University, I attempted to use the visualization tools offered by Tableau to bring visibility to infrequently accessed administrative records within the Pratt Archives’ collections and understand who lived in the Pratt Townhouses and what does that say about the institution and the surrounding community.

Data

The data utilized for these visualizations came from Lease Record Books and Leases that are currently held within the Buildings and Grounds Collection at the Pratt Institute Archives. Dating back to the early 1900s, these notebooks and documents are in fair-to-poor condition and are rarely requested despite the interest of many researchers in the development of Clinton-Hill and the larger Bedford-Stuyvesant neighborhoods. Within the ledgers and leases, one can find the names of the residents of multiple properties owned by Pratt Institute, the years they lived there, and the rent that they paid from 1905 up through 1972. For these visualizations, the focus is upon the twenty-seven properties that ran along Emerson Place, Steuben Street, and Willoughby Ave.

The initial collection and organization of the dataset was performed by a graduate student and researcher, Patrick Kingchatchaval, for a project he created in 2024. As part of his work, he pulled information from the ledger books and leases and input it into a spreadsheet, which he then transformed by the inclusion of color-coded rows and columns to enhance its aesthetic elements. Inspired by this work and motivated by the projects at California State University, University of Illinois, and Princeton, I sought to further explore the visualization possibilities for this dataset as well as expand into interactive mapping. 

To prepare the data, I reviewed the original documents to get an idea of their structure as well as review how they were aggregated. I then returned to the original spreadsheet to clean the data for import into Tableau for visualization. Cleaning the data involved: stripping the spreadsheet of various aesthetic elements such as row colors, font choices, and merged rows; normalizing the variables by removing dollar signs and changing the date ranges; and transposing data from a wide output to a long one using OpenRefine. Once I had performed these steps, I exported the file as a csv, which I imported into Tableau to began to develop the visualizations.   

First Exploration

Considering the sixty-five year time span of the data and the significant global events (world wars, epidemics, economic instability) and the local changes (economic shifts, housing policies changes, the intervention of Robert Moses) that occurred in that time period, I was interested in the continuity of tenancy of the faculty housing. Were they vacant for large stretches of time? Were they demolished, rebuilt, expanded? A visualization style which I believe reflects a sense of progression and continuity is that of the Gantt chart. Within the chart, data is first organized by the individual addresses and then proceeds to chart the individuals who lived there and the length of their stay. Initially, the individuals associated with each residence were arranged alphabetically, but I reordered so that they were arranged in order of their tenancy (first tenant in the building at the top of the list followed by the second successive tenant beneath). I felt that this created a flowing, cascading effect as well as allowed viewers to more clearly identify the progression of tenants. 

Second Exploration

Seeing trends already emerging from the Gantt chart visualization, I shifted from focusing upon the address and physical location to the individuals living within them. To do so, I developed a bar chart reflecting the number of years individuals lived in faculty housing in total. Within the bar chart, I directly mapped the name of the individual to a count of the total number of years of their residence, which was represented by a colored circle next to their name. The individual circle data points have a blue color gradient applied to them which reflects the specific years that the individuals have lived within the faculty housing. For this initial phase, I elected to not include labels for each row as I thought the existing blue circles representing the year count sufficiently for the viewers and easily identified those individuals with the longest tenure as residents of the Pratt Townhouses.

Third Exploration

Given the ongoing scholarship and investigation into rent fluctuations, demographic changes, and gentrification, I decided to include a visualization that reflected the rent data within the dataset. To do so, I generated a line graph whose data points represent the contracted rental rate for residents each year from 1907 to 1972 for each residential building. I chose the line graph to allow users to perceive the general trends over time as well as view any dynamic changes. Keeping to the consistent color palette previously introduced, the line depicting the datapoints is a shade of blue.

Findings

Looking at these initial visualizations, I believe that there are interesting trends and insights to be gained even at this initial stage of the project. Despite the broad time-span within a city that experienced various periods of economic decline and rebirth as well as cycles of rapid development and deterioration, Pratt was able to maintain surprisingly steady housing for its faculty and staff for the vast majority of the time documented within the ledgers. While not necessarily explaining what factors fed into this position, this continued presence potentially reflects an insulation that the Institute enjoyed from the rest of the city’s fluctuating financial status. 

Another interesting data point reflected in the visualizations is the length of time that some individuals resided within the housing reserved for Pratty faculty and their family. In multiple instances, there are individuals such as Isabel Ely Lord, William V. Gorham, Fannie Oakes, Edmund P. Davis, Beulah E. Stevenson, and others who lived in one particular address for over thirty years. This length of stay points to the importance and relevance of these individuals to the history of the institution and its work. The number of years alone could signal to researchers that it is worth investigating the impact that these individuals may have had on their departments or offices. With such longevity, they could have exerted an influence on the course of programs, pedagogy, facility operations, and more. It may also be worth exploring their interaction with the neighborhood organizations and local art circles due to their positions as instructors and practicing artists living and working in Brooklyn. 

A third trend or insight from the examples of data visualization comes from the charting of the evolution of the rent prices over time. It is yet another reflection of the ability of the Pratt administration to maintain a fairly stable financial burden for their employees despite broader and local changes. On average the majority of the rent began at around $40 and ended beneath $200, over sixty years later. Exploring other data to contextualize these rent prices versus other residences in the area and the broader city would be even more instructive, however, the visual changes of the line graph suggest a stability that I suspect was not experienced across the board in New York City. 

Initial Reflections

One way in which these initial visualizations can be improved would be to come up with a better way to reflect the passage of time than the existing color gradient in use. While the color gradient does convey an increase, an intensification of the color does not necessarily correspond conceptually to an increase in the amount of years. The consistent use of the blue color and the gradient across two visualizations might help orient the audience to its significance, but there is most likely another method that could be more intuitive and better representative of the time element of the visualizations. 

Another issue is the volume of residents overall within the dataset. In the report for this project, I had to resize and crop portions of the visualizations to avoid overcrowding the pages. As discussed before, the inclusion of all of the residents and calling the individual number of years provides important insight regarding the institution, the neighborhood, and more and so I am hesitant to alter the dataset. Given that the inclusion of all of these individuals was causing legibility and sharing issues, I will be exploring additional methods such as a radial graph to attempt to maintain the individual focus while reducing the footprint of the visualization.

Another analytical element that could increase the impact and weight of rent data would be to calculate and visualize the rate of change of the rent increase as opposed to the pure figures. The rate of change would likely better reflect the experience of the individuals as well as contextualize the data across time. Even though the audience knows that times are different and the valuation of the dollar has significantly changed, a $1 change in monthly rent versus a 10% increase in rent is understood differently. Locating and adding in a city average (if possible) to the visualization would provide even more insight as a contrast to the experience of the individuals living in the Pratt Townhouses. It could be even more effective to average all of the rent prices of the individual faculty housing units and compare that to a city average in a line graph. 

Future Directions

The first step would be to test and implement the changes I have suggested within the previous section. I will also submit it for feedback to users who better represent the target audience of researchers as well as the general public who might be interested in our collections, but are intimidated by or unfamiliar with archival research. 

After implementing those changes, the next step for the project would be the integration of the data into an interactive map that would allow users to adjust the time and see the change (or lack thereof) in residents as well as the rent prices over the course of sixty-five years. Of the visualizations I have presented so far, I believe that this would provide the most engaging way for researchers or individuals interested in Pratt and the surrounding community to view and understand these data points. While likely more complex to implement, it would allow me to introduce a timeline element and allow for filters to be toggled on and off within one overarching visual element. Ultimately, I feel that this form of visualization would offer the best means to incorporate all the data into a single, dynamic experience.