Predicting Elections with Twitter: What 140 Characters Reveal about Political Sentiment
By Andranik Tumasjan, Timm Oliver Sprenger, Philipp G. Sandner, Isabell M. Welpe. 2010
2,952 citations recorded by the source corpus.
Twitter is a microblogging website where users read and write millions of short messages on a variety of topics every day. This study uses the context of the German federal election to investigate whether Twitter is used as a forum for political deliberation and whether online messages on Twitter validly mirror offline political sentiment. Using LIWC text analysis software, we conducted a content-analysis of over 100,000 messages containing a reference to either a political party or a politician. Our results show that Twitter is indeed used extensively for political deliberation. We find that the mere number of messages mentioning a party reflects the election result. Moreover, joint mentions of two parties are in line with real world political ties and coalitions. An analysis of the tweets’ political sentiment demonstrates close correspondence to the parties' and politicians’ political positions indicating that the content of Twitter messages plausibly reflects the offline political landscape. We discuss the use of microblogging message content as a valid indicator of political sentiment and derive suggestions for further research.
Subfields: Artificial Intelligence, Communication, Sociology and Political Science
How researchers see this paper
The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.
In Artificial Intelligence
- Saif M. Mohammad
- Mohit Bansal
- Yejin Choi
- Jianfeng Gao
- Yoshua Bengio
- Noah A. Smith
- Percy Liang
- Luke S. Zettlemoyer
In Communication
- Derek Ruths
- Sarah Vieweg
- Daniel Gayo-Avello
- Filippo Menczer
- Alessandro Flammini
- Davood Rafiei
- Aibek Makazhanov
- Leysia Palen
In Sociology and Political Science
- Filippo Menczer
- Alessandro Flammini
- Daniel Gayo-Avello
- Krishna P. Gummadi
- Xinyi Zhou
- Zachary C. Steinert-Threlkeld
- Jay J. Van Bavel
- Huan Liu
Citations in the corpus
- Truthy: mapping the spread of astroturf in microblog streams (2010, background): “, predictions of box office success [4] and the results of political elections [48].”
- Predicting political preference of Twitter users (2013, background): “Happy birthday to Her Majesty, Queen Elizabeth II, Queen of Canada.”
- The Demographics of Social Media Users in the Russian-Language Internet (2017, methodology): “…data has been widely used for quantitative measurement and prediction of real-world events including prediction of the price level in the stock market [3], sentiment analysis of brands [21], prediction of movie revenues [1], earthquake prediction [20], and prediction of election outcomes [14] [27].”
- Comparing and combining sentiment analysis methods (2013, methodology): “There are multiple methods for measuring sentiments, including lexical-based approaches and supervised machine learning methods.”
- SemEval-2014 Task 9: Sentiment Analysis in Twitter (2014, background): “automatically mine and understand people’s opinions and sentiments has only recently been the subject of research (Jansen et al., 2009; Barbosa and Feng, 2010; Bifet et al., 2011; Davidov et al., 2010; O’Connor et al., 2010; Pak and Paroubek, 2010; Tumasjan et al., 2010; Kouloumpis et al., 2011).”
- Studying User Income through Language, Behaviour and Affect in Social Media (2015, background): “The vast amount of publicly available user-generated content on social media enables the study of complex problems for which sufficient data was not available before in various domains, such as health, politics or economics [1–3].”
- Studying political microblogging: Twitter users in the 2010 Swedish election campaign (2012, background): “This distribution largely follows the findings of previous research (Honeycutt and Herring, 2009; Kwak et al., 2010; Tumasjan et al., 2010), and indicates that Twitter is used mainly for sending undirected messages, with its potential for conversation and dissemination employed at a much smaller rat”
- Gender Inference of Twitter Users in Non-English Contexts (2013, background): “Outside of English-based analysis, some attention has been given to European and East Asian elections, e.g. (Tumasjan et al., 2010; Giglietto, 2012; Kim and Park, 2012).”
- Using Hashtags to Capture Fine Emotion Categories from Tweets (2015, background): “Tumasjan et al. (2010) studied Twitter as a forum for political deliberation.”
- Natural Language Processing to the Rescue? Extracting "Situational Awareness" Tweets During Mass Emergency (2011, background): “, Bruce and Wiebe, 1999), with some recent research correlating sentiment classification with real-world behavior (Gilbert and Karahalios 2010; Tumasjan et al., 2010).”
- Limits of Electoral Predictions Using Twitter (2011, methodology): “Twitter volume data were supposedly able to predict electoral results in Germany in 2009 with amazing accuracy (Tumasjan et al. 2010), yet more elaborated methods did not correlate with pre-electoral polls in the US (O'Connor et al. 2010).”
- Predicting the Political Alignment of Twitter Users (2011, background): “Social media data has proven valuable for predictions in various domains, including politics [18], [38], [39].”
- Predicting Social Unrest Events with Hidden Markov Models Using GDELT (2017, background): “Event prediction has been explored in a variety of applications, including elections [20, 21], disease outbreaks [22], stock market movements [23, 24], social unrest event prediction [11, 12, 25–31], movie earnings [23], crime [32], and failure prediction [33].”
- DeepSentiBank: Visual Sentiment Concept Classification with Deep Convolutional Neural Networks (2014, background): “Sentiment models have been demonstrated to be useful in various applications including human behavior prediction [8], business [26], and political science [34].”