As you know from my previous posts, I am working with data that I gathered on Urban Lumbees from the University of Florida’s Samuel Proctor Oral History Program. Using a free program called Voyant Tools, I attempted to perform a text analysis of my data. Unfortunately, the attempt was rather unsuccessful and I don’t think I got anything useful out of it. I ended up with a bunch of charts and visualization that I did not understand and did not make any sense. This may have happened simply because of the nature of my dataset. I was stuck with unhelpful terms that would take too long to clean through Voyant, such as interview ID numbers, and responses to yes or no questions. Therefore, in order to show how Voyant can be useful in some circumstances, I decided to perform a text analysis on an article published by the Washington Post Magazine entitled, “What Makes Someone Native American? One tribe’s long struggle for full recognition” by Lisa Rab. The purpose of using this analysis is to gain high quality information from the text that both visually summarize and identify what is important from the article. The first thing I produced using Voyant was a cirrus, which is a fancy name for a word cloud. Though it seems cliche, this word cloud tell us not only about the article, but the issues addressed in the article. Because the word cirrus was created by counting the frequency of terms used in the article, we know that the bigger words appeared more frequently throughout the text.
Without having to have read the article, we can tell that the article is about the Lumbee Tribe seeking federal recognition as Indians. When we look at the words in the cloud, it is also interesting to compare the terms to one another in order to identify their relevancy. For example, not how the word benefits is much larger than identity, blood, or members. Perhaps this tells us something significant about the issues at hand. Maybe this implies that the question over federal recognition has less to do with Indian identity but has more to do with the issue of becoming eligible to receive benefits. Though I don’t encourage drawing hard conclusions about any text based on word clouds, I do think it is an interesting way to gain a new perspective on a text and can be useful in formulating new research questions.
I also encourage you to read the Washington Post article by the way. You can access it by clicking here.
