Small Town or Metropolis? Analyzing the Relationship between Population Size and Language

Proceedings of the 12th Conference on Language Resources and Evaluation (LREC 2020), pages 6287–6291 Marseille, 11–16 May 2020 c⃝ European Language Resources Association (ELRA), licensed under CC-BY-NC 6287 Small Town or Metropolis? Analyzing the Relationship between Population Size and Language Amy Rechkemmer♣, Steven R. Wilson♦, Rada Mihalcea♠ ♣Purdue University ♦University of Edinburgh, ♠University of Michigan arechke@purdue.edu, steven.wilson@ed.ac.uk, mihalcea@umich.edu Abstract The variance in language used by different cultures has been a topic of study for researchers in linguistics and psychology, but often times, language is compared across multiple countries in order to show a difference in culture. As a geographically large country that is diverse in population in terms of the background and experiences of its citizens, the U.S. also contains cultural differences within its own borders. Using a set of over 2 million posts from distinct Twitter users around the country dating back as far as 2014, we ask the following question: is there a difference in how Americans express themselves online depending on whether they reside in an urban or rural area? We categorize Twitter users as either urban or rural and identify ideas and language that are more commonly expressed in tweets written by one population over the other. We take this further by analyzing how the language from specific cities of the U.S. compares to the language of other cities and by training predictive models to predict whether a user is from an urban or rural area. We publicly release the tweet and user IDs that can be used to reconstruct the dataset for future studies in this direction. Keywords: population, cities, culture, social media 1. Introduction In recent years, online profiles and posts on social me- dia platforms have provided a wealth of information for researchers to analyze, leading to discoveries about the re- lationships between the words that people write online and their emotions (Suttles and Ide, 2013), demographics (Rao et al., 2010), values (Boyd et al., 2015), geographic loca- tions (Han et al., 2014), and more. Through these linguistic associations, we can not only predict certain characteristics about the authors of these messages, but also get a glimpse of how people of different demographics interact with their world and process information. Prior studies have analyzed how location affects the type of language that people use, often looking at text written by authors from different countries when exploring cross- cultural differences (Poblete et al., 2011; Garcia-Gavilanes et al., 2013). However, it is not always necessary to look at multiple countries in order to view different cultures. Re- cent electoral results in the United States exemplify a divide in the political opinions between those living in densely populated areas and those living in rural parts of the coun- try (Scala and Johnson, 2017). This urban-rural ideological divide is intensifying, yet the people living in these areas depend on one another more and more (Lichter and Ziliak, 2017), interacting at the geographical borders, but also on- line through platforms like Twitter. Can we leverage text analysis tools …

Content truncated to 500 words