How do politicians use Facebook? An applied Social Observatory

Simon Caton, Margeret Hall, Christof Weinhardt

Research output: Contribution to journalArticle

14 Citations (Scopus)

Abstract

In the age of the digital generation, written public data is ubiquitous and acts as an outlet for today's society. Platforms like Facebook, Twitter, Google+ and LinkedIn have profoundly changed how we communicate and interact. They have enabled the establishment of and participation in digital communities as well as the representation, documentation and exploration of social behaviours, and had a disruptive effect on how we use the Internet. Such digital communications present scholars with a novel way to detect, observe, analyse and understand online communities over time. This article presents the formalization of a Social Observatory: a low latency method for the observation and measurement of social indicators within an online community. Our framework facilitates interdisciplinary research methodologies via tools for data acquisition and analysis in inductive and deductive settings. By focusing our Social Observatory on the public Facebook profiles of 187 federal German politicians we illustrate how we can analyse and measure sentiment, public opinion, and information discourse in advance of the federal elections. To this extent, we analysed 54,665 posts and 231,147 comments, creating a composite index of overall public sentiment and the underlying conceptual discussion themes. Our case study demonstrates the observation of communities at various resolutions: “zooming” in on specific subsets or communities as a whole. The results of the case study illustrate the ability to observe published sentiment and public dialogue as well as the difficulties associated with established methods within the field of sentiment analysis within short informal text.

Original languageEnglish (US)
JournalBig Data and Society
Volume2
Issue number2
DOIs
StatePublished - Jan 1 2015

Fingerprint

facebook
Observatories
politician
internet community
community
Data acquisition
social indicators
twitter
interdisciplinary research
data acquisition
formalization
Internet
social behavior
search engine
public opinion
documentation
Communication
Composite materials
communications
data analysis

Keywords

  • Facebook
  • Social Observatory
  • computational social science
  • political discourse
  • sentiment analysis
  • text analytics

ASJC Scopus subject areas

  • Computer Science Applications
  • Information Systems
  • Information Systems and Management
  • Communication
  • Library and Information Sciences

Cite this

How do politicians use Facebook? An applied Social Observatory. / Caton, Simon; Hall, Margeret; Weinhardt, Christof.

In: Big Data and Society, Vol. 2, No. 2, 01.01.2015.

Research output: Contribution to journalArticle

Caton, Simon ; Hall, Margeret ; Weinhardt, Christof. / How do politicians use Facebook? An applied Social Observatory. In: Big Data and Society. 2015 ; Vol. 2, No. 2.
@article{19443631d0cf4b6cb3275f64814b577c,
title = "How do politicians use Facebook? An applied Social Observatory",
abstract = "In the age of the digital generation, written public data is ubiquitous and acts as an outlet for today's society. Platforms like Facebook, Twitter, Google+ and LinkedIn have profoundly changed how we communicate and interact. They have enabled the establishment of and participation in digital communities as well as the representation, documentation and exploration of social behaviours, and had a disruptive effect on how we use the Internet. Such digital communications present scholars with a novel way to detect, observe, analyse and understand online communities over time. This article presents the formalization of a Social Observatory: a low latency method for the observation and measurement of social indicators within an online community. Our framework facilitates interdisciplinary research methodologies via tools for data acquisition and analysis in inductive and deductive settings. By focusing our Social Observatory on the public Facebook profiles of 187 federal German politicians we illustrate how we can analyse and measure sentiment, public opinion, and information discourse in advance of the federal elections. To this extent, we analysed 54,665 posts and 231,147 comments, creating a composite index of overall public sentiment and the underlying conceptual discussion themes. Our case study demonstrates the observation of communities at various resolutions: “zooming” in on specific subsets or communities as a whole. The results of the case study illustrate the ability to observe published sentiment and public dialogue as well as the difficulties associated with established methods within the field of sentiment analysis within short informal text.",
keywords = "Facebook, Social Observatory, computational social science, political discourse, sentiment analysis, text analytics",
author = "Simon Caton and Margeret Hall and Christof Weinhardt",
year = "2015",
month = "1",
day = "1",
doi = "10.1177/2053951715612822",
language = "English (US)",
volume = "2",
journal = "Big Data and Society",
issn = "2053-9517",
number = "2",

}

TY - JOUR

T1 - How do politicians use Facebook? An applied Social Observatory

AU - Caton, Simon

AU - Hall, Margeret

AU - Weinhardt, Christof

PY - 2015/1/1

Y1 - 2015/1/1

N2 - In the age of the digital generation, written public data is ubiquitous and acts as an outlet for today's society. Platforms like Facebook, Twitter, Google+ and LinkedIn have profoundly changed how we communicate and interact. They have enabled the establishment of and participation in digital communities as well as the representation, documentation and exploration of social behaviours, and had a disruptive effect on how we use the Internet. Such digital communications present scholars with a novel way to detect, observe, analyse and understand online communities over time. This article presents the formalization of a Social Observatory: a low latency method for the observation and measurement of social indicators within an online community. Our framework facilitates interdisciplinary research methodologies via tools for data acquisition and analysis in inductive and deductive settings. By focusing our Social Observatory on the public Facebook profiles of 187 federal German politicians we illustrate how we can analyse and measure sentiment, public opinion, and information discourse in advance of the federal elections. To this extent, we analysed 54,665 posts and 231,147 comments, creating a composite index of overall public sentiment and the underlying conceptual discussion themes. Our case study demonstrates the observation of communities at various resolutions: “zooming” in on specific subsets or communities as a whole. The results of the case study illustrate the ability to observe published sentiment and public dialogue as well as the difficulties associated with established methods within the field of sentiment analysis within short informal text.

AB - In the age of the digital generation, written public data is ubiquitous and acts as an outlet for today's society. Platforms like Facebook, Twitter, Google+ and LinkedIn have profoundly changed how we communicate and interact. They have enabled the establishment of and participation in digital communities as well as the representation, documentation and exploration of social behaviours, and had a disruptive effect on how we use the Internet. Such digital communications present scholars with a novel way to detect, observe, analyse and understand online communities over time. This article presents the formalization of a Social Observatory: a low latency method for the observation and measurement of social indicators within an online community. Our framework facilitates interdisciplinary research methodologies via tools for data acquisition and analysis in inductive and deductive settings. By focusing our Social Observatory on the public Facebook profiles of 187 federal German politicians we illustrate how we can analyse and measure sentiment, public opinion, and information discourse in advance of the federal elections. To this extent, we analysed 54,665 posts and 231,147 comments, creating a composite index of overall public sentiment and the underlying conceptual discussion themes. Our case study demonstrates the observation of communities at various resolutions: “zooming” in on specific subsets or communities as a whole. The results of the case study illustrate the ability to observe published sentiment and public dialogue as well as the difficulties associated with established methods within the field of sentiment analysis within short informal text.

KW - Facebook

KW - Social Observatory

KW - computational social science

KW - political discourse

KW - sentiment analysis

KW - text analytics

UR - http://www.scopus.com/inward/record.url?scp=84982316426&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84982316426&partnerID=8YFLogxK

U2 - 10.1177/2053951715612822

DO - 10.1177/2053951715612822

M3 - Article

AN - SCOPUS:84982316426

VL - 2

JO - Big Data and Society

JF - Big Data and Society

SN - 2053-9517

IS - 2

ER -