Introduction
For this project, I decided to create two “marker” maps on Flourish to visualize the birth and death locations of artists in the Tate dataset. The goal of this project was to visualize where artists came from to highlight disparities in racial and geographic representation.
Sources
I used a dataset from the Tate gallery that contained data of various artists born in the 1900s. There is data on their names, year of birth and death, place of birth and death, gender, and a URL that linked to a page that displayed each artist’s works and a short biography. To prepare the data, I used both Google Sheets and a Python script I wrote to clean the data.
Processes
This process was what took me the longest. I started by filtering out the columns that were not relevant to my analysis. I removed the id, gender, dates and URL columns. I then removed every artist that did not have a birth location and a death location since I wouldn’t be able to plot them on a map, and I also filled any empty cell with “None”. Once I had done that, I took my data over to Flourish, but I found that in order to plot points on the marker map, I needed to input latitudes and longitudes, so I decided to create a Python program that would do just that.
My Python program uses the geopy library to geocode a given location and return latitude and longitude coordinates. I won’t explain in depth what my program does line-by-line, but I will link it here. However, I will explain the problems I encountered when coding. I found that Python didn’t like reading special characters from the csv file, so I had to replace each special character with a general English letter (i.e. รถ to o). I also found that there was a slight issue with how the locations were put into the dataset, specifically U.S. locations. All the U.S. locations were inputted as follows: City, United States. This causes issues for cities that have duplicate names. For example, say you look up Austin on Google Maps. The first one that shows up is Austin, MN followed by Austin, TX. Say you want Austin, TX, but because Austin, MN is first, geopy will select Austin, MN and return its coordinates. Therefore, I had to find a way to alter the location to be city, state. Unfortunately, I couldn’t think of a quicker way, so I had to look up every artist that was born or died in the U.S. to alter the location. After doing all of this, my program was able to successfully retrieve the coordinates of each location and then write them onto a new csv file which I was then able to use with Flourish.
After cleaning my data, I used Flourish to visualize the data by creating marker maps where each marker is a location based on the coordinates given. I initially wanted to include both death and birth locations together, it got too visually cluttered, so I decided to split them up into two maps. The symbols for the markers conveyed its meaning, the baby marker represents location of birth and the cross represents location of death. When clicking on the marker, information about the artist’s name and year of birth/death is also displayed.
Presentation
I embedded the maps into this website to allow the user to be able to fully interact with the map by zooming in and out and click on the markers. The choice of an interactive map makes it easier on the user to understand the map rather than providing a static image of a cluttered map.
Significance
When looking at specifically the birth location map, we can see that there are an overwhelming number of markers on Europe and the United States. This highlights the focus on the Western world when it comes to art and artists. By visualizing this on a map, this trend becomes abundantly clear compared to only reading the raw csv file. This could potentially lead to further analysis on whether the artist’s born in Europe/United States were of European descent and raises questions of bias towards the Western world. Rather than simply presenting empirical findings, this project allows for analysis through a social-cultural lens to be able to see trends that otherwise might go unnoticed.