Blog Post

Having fun with Voyant tools

After learning how to use Voyant tools, I wanted to see whether it was able to reflect my qualitative research results.

Parts of my recent research are about how Korean ethnic media has reported the current issue of Korean diaspora in the United States. There was an agreement between the Korean and Japaense government about WWII victims, comfort women on December 28th 2015. After the agreement was made, there have been huge demonstrations, rallies, and campaignes among Koreans including Korean diapsora to denounce the agreement. In particular, it was considered as a critical issue for the Korean diaspora in the United States, because the contents of the agreement were going to be parts of the public schools' text book in California state. Discourses have been formulated on the basis of online discussions, and they led to the offline social movements. I looked into text boards of the online communities and Korean ethnic media webpages, and examined how their users presented their opinions and how these opinions are re-shaped in the public sphere.

With Voyant tools, I can clearly see what the key terms and ideas related to the topics are. The analysis is pretty much overlapped with my results from the qualitative textaul analysis. Also, different analysis based on the different time frames and different media show how the public opinions have been changed as time have passed. Even though it does not reveal any statistical significance, but only the word frequency, I still think that this is very effective tool to deliver the message when presenting.  Plus, I am so happy that it works with Korean texts. =)



Is there a separate Voyant page for Korean as with Japanese, or can you just paste the Korean into the main site? I imagine it's the latter since Korean has spaces whereas Japanese does not (and so required its own version), but just thought I'd ask. I'd like to advertise this to the Korean studies community at Penn.


Yay! I'm so excited to continue using this program with my narratives, once I convert them to text files :)


It makes sense that the results would be similar, but it alsmo makes me want to know more about the method used by the statistical analysis package (SAS?), and where it might be different from Voyant's word counts. I love your wordcloud attachement--first one I've seen in Korean!


I never realized I could use Voyant with Japanese until after I read your post and gave it a try. This is really going to help my Japanese studies. Thanks!