Comprehensive demographic dataset for Louisiana State University, Baton Rouge, LA, US including population statistics, household income, housing units, education levels, employment data, and transportation with year-over-year changes.

TigerDroppings.com, founded in 2001 by LSU alumnus Brian Fiegel, is among the most notable and active of any college sports forums on the internet, and its popularity hasn’t declined even as major social media platforms have come to dominate online spaces for discussion. The site’s userbase consists primarily of Louisiana residents and LSU graduates, though fans of other schools in the Southeastern Conference also frequent the site. The users on these sites fit a very specific demographic and they have little diversity. In a survey of users from numerous college sports forums, including TigerDroppings, it was found that 87% of users were male and 90% were white. Additionally, 76% had at least an undergraduate degree and 42% of users had a household income of $100,000 or greater. In November 2015, TigerDroppings had 129,244 users and now, seven years later, has 256,692 registered users; the site is still growing as fast as it did in the 2000s. There is a reason for this — these very specific demographics of the userbase are able to communicate in a way they otherwise couldn’t on Facebook or Twitter. Given these demographics, the forum takes on an overwhelmingly conservative tone in the opinions and sentiments regularly expressed. To put it simply, I couldn’t imagine a dataset that better encapsulates the psyche and mindset of white conservative men in Louisiana. Comprising almost 14 million political posts from 2014 to the present, it profiles the rise of Trumpism and the cataclysmic shifts seen in American politics in recent years.

Early in the site’s history, their off-topic board the “OT Lounge” was created and is the most popular board on the site, followed closely by their “Politics” board. Unlike many other similar forums, TigerDroppings relies solely on advertising to generate revenue, and all boards are free to view and create posts on. Only an email is required to sign up and all posts are anonymous; users are only outwardly identifiable by their chosen screennames. The functionality of the site has largely gone unchanged since its founding. Users can start a thread on a particular board and replies by other users are appended to the thread; there is no visible hierarchy to replies on threads, unlike platforms like Reddit, and it is very rudimentary by current standards. On every single reply in a thread there is an upvote and a downvote button; next to each button, their respective values are displayed, publicly showing the popularity of a user’s post. Users have been informally voting on political opinions and sentiments constantly, which I believe is rich for analyzing the rise of specific attitudes and rhetoric used among this demographic.

Attached to each post in the dataset are several pieces of metadata: upvotes, downvotes, username, date of post, date of thread creation, URLs from links contained in the post, URLs to images in the post, text from blockquotes, and the position of the post in its respective thread. Additionally, I was able to gather emails and phone numbers for approx. 3,000 users of the site through the Ticket Marketplace Board, as many users had posted contact info to interact with other users externally. Data from the OT-Lounge was able to be scrapped in its entirety from 2014 to present, though among data from the Politics Board there were some gaps. All data from 2015 was not publicly accessible for unclear reasons. But more interestingly, all threads from November 2, 2020 to January 7, 2021 — the day before the presidential election until the day after the insurrection at the U.S. Capitol — is not publicly accessible at all. I hypothesize there was significant activity talking about election fraud during that period, along with potentially incriminating information about posters who may have participated in the events on January 6th.

In terms of where I want to go from here with this dataset, I am interested in exploring if a model to isolate and predict political trends among this demographic could be feasible, along with exploring what potential uses it has as a tool for electoral politics in Louisiana. If anything, I want to do some anthropological research about this demographic that so clearly to me describes the social, cultural, and political environment I was raised in.