Analyzing social media, analyzing the social? A methodological discussion about the demoscopic and predictive potential of social media

被引：0
作者：
Santander P. [1 ]
Alfaro R. [1 ]
Allende-Cid H. [1 ]
Elórtegui C. [1 ]
González C. [1 ]
机构：
[1] Pontificia Universidad Católica de Valparaíso, Valparaiso
来源：
Quality & Quantity | 2020年 / 54卷 / 3期
关键词：
Chilean’s presidential election; Digital demoscopy; Election forecasting; Political communication; Social media;
D O I：
10.1007/s11135-020-00965-z
中图分类号：
学科分类号：
摘要：
The impact of computational technologies and the worldwide use of Internet entails a theoretical and methodological challenge for social scientists, considering the purpose of observing, interpreting and explaining human and social behaviour. Today, the digital environment seems to be an adequate space for this exploration and the emergence of the Web 2.0 offers common people the possibility of expressing and sharing their opinions on a daily basis. Due to the ubiquity of technology, Internet and social media in people’s lives, socialization and its expressiveness have changed. If this is the case, the means to measure the perceptions, opinions and judgements of citizens should also change. The immense quantity of data available to be analysed today poses a challenge for the traditional scientific model. In this sense, it could be necessary for social research to move towards the analysis of the web and consider the potential predictive capacity of digital demoscopy. A new field of study has opened, with interest in exploring the predictive capacity of social media in electoral contexts. As a research group comprised by linguists, communication experts and engineers we explored the predictive potential of social media in three national elections that took place in Chile during 2017. Our objective was to explore a methodological design that allows predicting the result of political elections through the use of inductive algorithms and the automatic processing of messages with political opinion in social media. Through computational intelligence, we were able to follow, collect and analyse millions of tweets, and to improve our forecast each time. Our learning based on empirical research was fundamental to improve our procedures and to refine our variables and, thus, improve our prediction. © 2020, Springer Nature B.V.
引用
收藏
页码：903 / 923
页数：20
相关论文
共 52 条
  • [1] Anderson C., The end of theory: The data deluge makes the scientific method obsolete, (2008)
  • [2] Aparaschivei P.A., The use of new media in electoral campaigns: analysis on the use of blogs, Facebook, Twitter and YouTube in the 2009 Romanian presidential campaign, J. Media Res., 2, 10, pp. 39-60, (2011)
  • [3] Aradau C., Blanke T., Politics of prediction: security and the time/space of governmentality in the age of big data, Eur. J. Soc. Theory, 20, 3, pp. 373-391, (2016)
  • [4] Barbera P., Less is More? How Demographic Sample Weights Can Improve Public Opinion Estimates Based on Twitter Data, (2015)
  • [5] Beauchamp N., Predicting and interpolating state- level polls using Twitter textual data, Am. J. Polit. Sci., 61, 2, pp. 490-503, (2017)
  • [6] Benoit W., Sheafer T., Functional theory and political discourse: televised debates in Israel and the United States, Journal. Mass Commun. Q., 83, 2, pp. 281-297, (2006)
  • [7] Bermingham A., Smeaton A., On using Twitter to monitor political sentiment and predict election results, Proceedings of the Workshop on Sentiment Analysis Where AI Meets Psychology (SAAIP 2011). Chiang Mai, Thailand: Workshop at the International Joint Conference for Natural Language Processing, pp. 2-10, (2011)
  • [8] Burrows R., Savage M., After the crisis? Big data and the methodological challenges of empirical sociology, Big Data Soc., (2014)
  • [9] Ceron A., Curini L., Iacus S., Politics and Big Data, (2017)
  • [10] Ceron A., Curini L., Iacus S.M., Porro G., Every tweet counts? How sentiment analysis of social media can improve our knowledge of citizens’ political preferences with an application to Italy and France, New Media Soc., 16, 2, pp. 340-358, (2014)
123456