Thesis Proposal

 For my senior thesis project, I wanted to see if there were any differences in the type of comments on different YouTube reactor’s videos. To start off, I had to understand more about YouTube. Youtube is a website that allows registered users to upload their own video content and interact with followers of their channel. This platform has sixteen main types of videos and over two billion active users. The creators of this website designed it to have different functions to interact with comments. One of the main functions that it has is that it gives the uploader the option to turn off the comments. Users can also “rate” comments using the thumbs-up or the thumbs-down buttons. (Madden et al., 2013, p.2) The comments with the most thumbs-up will be moved to the top of the comment section. While the comment section is mainly used for expressing opinions about the video, there are comments that have nothing to do with the video.

The second thing that I have to understand is about reaction videos and why people like them. According to Erich Alan Werner, reaction videos can be defined as “a vlog that captures the reaction of an individual (or group of individuals) as they encounter a particular text, usually a short media clip.” (2012, p.68) When putting out a reaction video, people are looking for the facial expressions and emotions that the creator is using. Emotion has a lot to do with communication, and Patricia G. Lange, an anthropologist, describes the face and its emotional expressivity as the key to vlogging’s power and popularity (2012, p.70) People feel like they have more of a connection with a face than they do with text. 

The final thing I have to understand is more about YouTube’s comment section. YouTube’s comment section is different from others because there is limited networking. Networking on YouTube is not as big as the networking you can find on Instagram or Twitter. It is harder to keep up with and participate in threads. Even though it is harder to connect with other people, YouTube has one of the most hostile comment sections with many hostile comments. YouTube has tried to monitor the comment section on videos, but due to the large number of users and their activity on the platform, it failed to differentiate hostile spaces on the site.

Because I am focusing my project on seeing if there is a difference between reactors’ comment sections on YouTube, I had to start looking for a song choice and YouTube reactors. The song I chose is Driver’s License by Olivia Rodrigo. I chose this song because it is the most popular song of 2021. It was number one on the U.S. Billboard chart for eight weeks and number one on Billboard’s Global Excluding U.S. chart for nine weeks. Then I had to decide which reactor's comment section I wanted to analyze. For this project I decided to choose two different types of reactors. The first type of reactor I chose was just a general reactor. What I describe as a general reactor is just normal people listening to the song for the first time. To find the reactor that I chose, I went to YouTube, searched reaction videos to “Olivia Rodrigo Driver’s License reaction”, filtered the searches by most viewed video, and scrolled until I found the first general reactor. That is when I found the YouTube channel “Brother.” The second type of reactor that I wanted to compare the general reactor to was a reactor who was a poet. When I searched for reactors who were poets, I only found one channel. The channel that I found is called “Sher the Truth.” I went to each of their comment sections and filtered the comments so all of the top comments would be at the top of the section. I then copied the top twenty comments and pasted them in a Google Doc. To analyze these comments, I started out by looking to see if they had five components in them: correct capitalization, correct punctuation, correct spelling, full words/phrases, and use of emojis. I chose these five components because I think they are the bases of a well written sentence and/or paragraph. In the Doc, I went and highlighted every component I found in a different color. I then put all of my data into an excel sheet. After that, I included five more components that I would analyze. I found these components from an article called “Sifting useful comments from Flickr Commons and YouTube” by Elaheh Momeni, Bernhard Haslhofer, Ke Tao, and Geert-Jan Houben. The five components are: named entity types variety, the average length of sentences, topical conformity, subjectivity tone, and tone. I chose these five out of a list they had because they either required no program/equation or I could modify them to fit what I am trying to do. I analyzed data the same way I did for the previous data analysis.

As I analyzed the data that I gathered, I noticed that there were not very many differences between both reactors. I saw both creators had three categories where they had the same amount of problems as each other. I was shocked to see that there were no differences because I believed that the comments under the Sher the Truth's video would have been better written than the comments on Brother's video. Moving forward, I plan to include different types of reactors, different songs, different things to look for in the comments, and maybe videos' that react to full length albums. By including more creators and changing some things, I hope to find some actual differences between the comment sections under each reactor.