In previous blog posts, I mentioned that I was in the midst of coding and identifying variables for individual posts on the “Long Beach, NY Hurricane Sandy Information” Facebook page. Now that I have completed this component of my research, I want to share my results!
At one of our first research meetings, I was assigned to take snapshots of posts from the end of October through the end of November. Even though I tried to predict certain patterns that would exist during these months, my data has proven to be far more informative than any of my tentative hypotheses.
As I played with my data, I decided that it would be important to determine how many posts there were for each genre.
Information-Seeking and Information Sharing are the most popular genres in the Facebook group for the October and November months. This conclusion did not particularly surprise me because I has assumed that group was initially used as a way to offer services and goods as well as to collaborate updates regarding the recovery process. While I acknowledge that the Facebook group also serves as an outlet for Long Beach residents, post-residents, and people from across the country and world to unleash their emotions, the posts from October and November pertain most to information sharing and seeking. I was confused as to why Information-Seeking constitutes 28% of the posts whereas Information-Sharing constitutes 53% of the posts. I then remembered that during my visit to Long Beach, I spoke to a woman who actively used the Facebook group when she evacuated Long Beach after Hurricane Sandy hit. She explained that she was constantly checking the Facebook group on her computer to provide her friends in Long Beach with updates regarding the locations of aid, donations, and events. Thus, the discrepancy between Information-Seeking and Information-Sharing is probably correlated to the lack of access that Long Beach residents had to electricity, especially to computers and internet, following Hurricane Sandy.
I then decided to look at the average number of likes, comments, and shares per genre to try to determine which posts were the most popular. In my graph, the green line represents the average number of likes, the red line represents the average number of shares, and the blue line represents the average number of comments.
I concluded that the genres, except for Information-Seeking, had more likes than comments or shares. Solidarity/Community and Memory had the greatest number of likes, which alludes to the power of a “like” on Facebook. The “likes” regarding Solidarity/Community and Memory reflect the cohesiveness of Long Beach as well as external support. I was intrigued as to why Information-Seeking posts seemed to have the fewest comments. However, I realized that the purpose of Information-Seeking posts is to elicit answers, not necessarily likes! I was not surprised to learn that posts about Information-Sharing had the most “shares” because a “share” on Facebook is a way to publicize information to a bigger audience. Therefore, the Facebook users who shared these posts most likely recognized the posts’ potential impact and applicability. The number of comments seems to appear fairly consistent with slight increases in Contemporary Documentation, Info-sharing, Memory, and Solidarity/Community. I wonder why it is more challenging to track changes in the average number of comments per genre.
I made another graph to determine which gender or admin/other were posting most frequently for each genre. Before we had any quantitative data, we mentioned that we had been seeing an overwhelming number of female posters. I wrote a blog post about the apparent gender division of labor to express some of the observations that I have been seeing regarding gender. However, this graph makes me want to know more! Why are women posting more than men in almost every genre? What is the significance of gendered Facebook use?
I am really excited to further analyze and compare our data. I feel like our data gives us a better understanding of the observations and hypotheses that we made. We just need to keep asking WHY we are seeing particular trends in the data!