For my midterm project, I decided to create a map displaying where female artists found in the Tate gallery are from. The corresponding repository contains information on a vast number of artists from the 1900’s, but narrowing down the data to only female artists helped make the data easier to work with. I did this by using OpenRefine’s ‘text filter’ feature to target the ‘gender’ column. This brought down the file to about 100 artists.
After exporting this file as a csv, I was able to then delete unnecessary columns that weren’t relevant to the information I was plotting. I imported the spreadsheet to Palladio to create the map, but then realized that in order for the birth places and names to correspond accurately, I would need to make more spreadsheets to connect variables together. In a separate spreadsheet, I used the GeoCode extension to get the longitude and latitude of the artists’ birth places. This created two new columns, so I formatted them as one column to better suit how Palladio reads coordinates. In another spreadsheet, I used more Google Spreadsheet formulas/formatting settings to find the unique set of countries where these artists’ are from. From there I was able to create another column to hold the frequency of artists who were born in a certain country.
This is the csv that I started out with after using OpenRefine:
These are the spreadsheets I had to import into Palladio to create the graph:
Palladio currently doesn’t have the option to embed visualizations, but a screenshot of the map can be found above. The red dots indicate the 100 female artists, and upon scrolling over those dots, users can see their name. The blue dots represent the unique countries which the artists are from. I used the point size feature so that larger dots represent countries where more artists are from. This allows users to gain the insight that female artists in the Tate Gallery are primarily from the United States or the United Kingdom. Users can also find artists given that they have a particular location in mind. For example, if I wanted to know about an artist from Canada, I would look in that area for any red dots.
These insights aren’t terribly remarkable and the map itself is very simple, but creating the spreadsheets and using Palladio as a whole took a fair amount of trial and error. One problem I had with Palladio was getting different parameters from different spreadsheets to essentially ‘sync up’. This took some time to get used to so I found myself making and remaking different spreadsheets with the same types of information. Another problem I had with Palladio is that moving from different tabs caused the map layers I made to get deleted, but this could be solved by improving my literacy of network analysis softwares as a whole. In the future, I’d like to look into data sets that have more information about its subjects aside from place of birth/death, and try overlaying different graphs together.
