CartoDB, a web-based tool, allows the user to connect data with its software in order to produce mapping results. With its drag and drop functions, it “enables you to predict key insights from your location data” (Carto Builder). It allows the user to include several datasets. When they are ready to be published, the user can share them by embedding the link or sharing the URL. There are two main views: Data and Map. Also, CartoDB allows the user to select fields from the slide menu. There are different types of maps for visualization. You may select each map and adjust the speed and selected data. If the user wants to Visualize the data, he or she will select Visualize. The image can also be exported and saved. A title can be added by selecting elements. Also, text and images can be added to the map. Another key interesting feature is the layered map.
For my CartoDB activity, I used a dataset that was given by Dr. Robertson. The dataset is from the WPA Slave Narratives Collection, which belongs to the Library of Congress. Setting up a CartoDB account was easy. In addition to the “Sign Up for Free” option, there are options for monthly subscriptions, which allow more data storage and sharing. Fortunately, I was glad to have the free option. After I added the URL of the dataset to CartoDB, I was able to view the dataset. By selecting the Map at the top of the screen, I saw a map with orange points. Each point had an information window that can be seen by clicking on it. In order for the fields of data to be activated, I had to select the fields in the slide menu. After numerous attempts, I learned to simply click on the grey area by the buttons next to the field in order to activate it. Without Dr. Robertson’s advice, I would have continuously failed by sliding the button to the right to activate it.
After the fields of age, sex, location, etc. were selected, I was able to save the map and publish it. I also exported it. When I clicked on one of the orange points, I was able to see the selected fields with information. At different points on the map, it reveals that the interviews were mostly conducted throughout Alabama. Most of the interviews were near Birmingham. However, they seem to be conducted away from the major cities and closer to the rural areas. The map also reveals that there were more interviews conducted southwest of Birmingham, below the city of Tuscaloosa. After viewing each point on the map, I noticed that the ages of the former slaves ranged from 78 to 112. The “Type of Slaves” were mostly unknown. Some of them were either Field or House slaves. Not all of the former slaves were born in Alabama; some of them came from South Carolina or Georgia. The dates of the interviews ranged from late May to early July during the summer of 1937. The map also reveals that the interviews were conducted in areas that were close to the major highways that exist today but probably not in 1937. In 1937, interviewers probably had challenges with travel and weather in order to meet the interviewees, and the interviewees mostly lived in areas of familiarity and segregation. Also, the map shows the racial divide and isolation of African Americans and former slaves in the Deep South.
After viewing the Simple/Point map, I viewed five more types of maps. The Torque map reveals the location of each interview at a certain date and with animation, it shows change over time. This map shows that the first two interviews were conducted near Birmingham and Meridian. Then, the interviews were conducted south and north of Birmingham. Some of the points overlapped. I wonder if the interviews were conducted twice by two different interviewers or by the same interviewer. In comparison to the simple point map, this Torque map provides a visual reference of time and movement (location change) of how the interviews were conducted.
After viewing the Torque map, I selected the Cluster and Intensity maps. The Cluster and Intensity maps reveal the number of interviews conducted in a certain location. In comparison to the simple map and torque map, it shows the area/location that had a higher number of interviews and which areas were the hot spots. 10 or more interviews were near Meridian (12), in Birmingham and close to Montgomery (18). These were the concentrated areas of the interviews.
Next, I viewed the Heat map. The Heat map shows a better visual reference for the number of interviews in a cluster. In comparison to the Point map and Cluster map, the Heat map shows the concentration of the greater number of recorded interviews close to and in Montgomery. Birmingham and areas close to Merdian show a good number of interviews. The areas north of Birmingham show a lighter blue color, indicating less number of recorded interviews. The Heat map provides a closer look at the number of recorded interviews.
Then, I viewed the Category and Torque maps. The Torque map is a type of Category map but with animation to show the change over time. The Category maps reveal which type of slave was interviewed in each area. Most of the slaves who were interviewed were “unknown” for type of slave. In comparison to the point map and cluster map, the Category map reveals a specific area on the map and key for the type of slave. It gives the user a better visual reference and idea of how many areas on the map were actually unknown types rather than field or house slaves. The areas of higher number of recorded interviews had a higher number of unknown categories.
In addition to viewing each map, CartoDB allows the user to layer the maps. For my activity, the layered map reveals the U.S map (mostly to the east and south), states, cities, and specific locations in the cities of where the interviews were conducted. It shows that interviews were conducted in states other than Alabama. It includes interviews in Kentucky, Virginia, North Carolina, and Kentucky. Each orange point reveals an information window with age, the names of the interviewee and interviewer, the latitude and longitude, sex, date, type of slave, location of interview, place of birth, etc. It also shows the area for the types of slaves who were interviewed for each location point. My layered map is a combination of the point and category maps with added elements of title and image.
As a geospatial tool, CartoDB invites the user to examine and analyze data by providing visual representations of datasets. It transforms text into visual translations. Each map provides a different way of looking at the data on a map. For my CartoDB activity, I began to ask questions about the interviews by visualizing the various maps. What information is missing? What does the change over time signify? In some of the interviews, why is the type of slave indicated as “unknown.” I wonder if the interviewer decided not to ask the question or if the information was confidential. I also noticed the segregation or racial divide on the maps by viewing the interview locations and the lowest to highest number of interviews. The personal information about each former slave in connection to the location point of the interview on the map provided a humanistic approach. From my point of view, the data became more subjective than objective when CartoDB transformed the datasets into maps.
