Social Science Computer Review

Volume 32, No. 4

April 2014


Table of Contents


Symposium Issue on Quantifying Politics Using Online Data


Guest Editors: Yelena Mejova & Ingmar Weber


Introduction to the Symposium Issue on Quantifying Politics Using Online Data /  Yelena Mejova & Ingmar Weber


Measuring Moral Rhetoric in Text / Eyal Sagi & Morteza Dehghani

In this paper we present a computational text analysis technique for measuring the moral loading of concepts as they are used in a corpus. This method is especially useful for the study of online corpora as it allows for the rapid analysis of moral rhetoric in texts such as blogs and tweets as events unfold. We use latent semantic analysis to compute the semantic similarity between concepts and moral keywords taken from the “Moral Foundation Dictionary”. This measure of semantic similarity represents the loading of these concepts on the five moral dimensions identified by moral foundation theory. We demonstrate the efficacy of this method using three different concepts and corpora.

Promoting Civil Discourse through Search Engine Diversity / Elad Yom-Tov, Susan Dumais & Qi Guo

Abstract: The ability of modern web services such as news aggregators and search engines to tailor their results to the tastes of individuals, together with people’s preference for reading opinions which reinforce their own viewpoints, have raised concerns that people are nowadays only exposed to a narrow range of view-points, a phenomenon referred to as the ”filter bubble”. Several attempts have been made to nudge people into reading opposing views, mostly on political topics. These attempts focused on presentation of the opposing views in ways which would encourage people to read diverse opinions, but their success was marginal.  In this paper we focus on exposure to varied political opinions with a goal of improving civil discourse. We develop a method to algorithmically encourage people to read diverse political opinions and test it when people actively seek information. First, by analyzing data from a popular search engine we show that people are indeed more likely to read opinions consistent with their own: While 76-81% of people read pages from highly polarized news outlets consistent with their own views, only 4-6% read pages from the most polarized opposing sites. Interestingly, it is more likely that they will read news from opposing sites when the language model of a particular news item is close to the language model of their own political leaning.


Coevolution of Political Discussion and Common Ground in Web Discussion Forum / Hai Liang

Abstract: Common ground is vital for developing deliberative democracy. The current study employs text mining techniques to measure common ground in online political discussions and examines how the structure of political discussions coevolves with common ground over time. The present study collected 175,960 messages over a period of 13 months, from a popular discussion forum on 2012 U.S. Presidential Election. Common ground is measured by a semantic similarity network and an interpretive framework network. The former emphasizes shared political knowledge, while the latter emphasizes shared interpretations. In addition, this study explores the coevolutionary process of political discussion and common ground. Results were obtained by employing longitudinal network analysis. They suggest that political discussions could facilitate the achievement of common ground that might further serve as a facilitator of political discussion among the participants.


The Life and Death of Political News: Measuring the Impact of the Audience Agenda Using Online Data / Jonathan Bright & Tom Nicholls

Abstract: The rapid development of online media as a major location for news consumption has stimulated a variety of debates about how journalism is changing in the internet era. Of particular importance have been worries about a potential turn towards populism, whereby journalists and editors shift away from reporting what is newsworthy to what their audience wants to hear, supported by the widespread availability of audience metrics. A wealth of ethnographic research has pointed to the potential importance of such statistics; but little quantitative work has been conducted to test for the existence of a relationship between audience behavior and editorial decisions. This study seeks to fill that gap. Based on a novel dataset of over 40,000 articles published in 5 major UK news outlets over a period of 6 weeks, we explore the relationship between a news story’s readership and its likelihood of being removed from the front page, based on the ‘most read’ lists common to many news websites. We find that being a ‘most read’ article decreased the short-term likelihood of being removed from the front page by around 25%, and that this effect was broadly similar for both political and entertainment news. Surprisingly, we find a considerably greater influence in ‘quality’ publications than their tabloid counterparts. Our results are discussed as evidence of a still limited, but potentially developing, turn towards online populism.


Driving Regulation: Using Topic Models to Examine Political Contention in the U.S. Trucking Industry / Karen Levy & Michael Franklin

Abstract: Public comments submitted during agency rulemakings can provide rich insight into stakeholders’ viewpoints around contentious political issues, but have been largely untapped as a data source by social scientists. This is in part due to the lack of access to comments in machine-readable formats, and in part due to the difficulty of analyzing large corpora of textual data. However, new online repositories and analytic methodologies are beginning to open up this trove of data for researchers. Using data from the online portal, we employ probabilistic topic modeling to identify latent themes in a series of regulatory debates about electronic monitoring in the United States trucking industry. Our model suggests that different types of commenters use alternative discursive frames in talking about monitoring. Comments submitted by individuals were more likely to place the electronic monitoring debate in the context of broader logistical problems plaguing the industry, such as long wait times at shippers’ terminals, while organizational stakeholders were more likely than individuals to frame their comments in terms of technological standards and language suggesting cost/benefit quantification.


Linguistic Factors Associated with Propagation of Political Opinions in Twitter / Kirill Dyagilev & Elad Yom-Tov

Abstract: In the last decade, social networks have emerged as a significant media platform for discussion and dissemination of political information. Pew surveys (Smith, 2011) found that 22% of adult Internet users participated in political campaigns through at least one of the major social media platforms during the 2010 US elections. In this paper, we attempt to characterize linguistic factors that affect the dissemination of a political message on Twitter. This is important because the characterization of messages that were posted or retweeted by users from both political camps can provide a unique insight into how rival audiences can be reached. We analyze about 20,000 Twitter users that expressed explicit support for Barack Obama or Mitt Romney during the 2012 US Presidential election. We extracted approximately 344,000 tweets by these users that contained links to political web pages and that appeared in the three months prior to the election. We show that the language used in a linked page is correlated with page's popularity in terms of the total number of users that posted a link to this page. Pages that are written in a more conservative language are posted by a larger number of users than liberal-leaning pages. We also observe that pages that use a highly polarized language are generally posted or retweeted by users from a single political camp. Pages that are posted by both Romney and Obama supporters are usually written in a relatively neutral language.  In addition, when the retweeting user is Republican, the text accompanying the retweeted page is likely to be modified, most likely to interpret the link according to the users’ own point of view.


International Relations: State-Driven and Citizen-Driven Networks / Andrew Crooks, David Masad, Arie Croitoru, Amy Cotnoir, Anthony Stefanidis & Jacek Radzikowski

Abstract: The international community can be viewed as a set of networks, manifested through various transnational activities. The availability of longitudinal datasets such as international arms trades and United Nations General Assembly (UNGA) allows for the study of state-driven interactions over time. In parallel to this top-down approach, the recent emergence of social media is fostering a bottom-up and citizen driven avenue for international relations (IR). The comparison of these two network types offers a new lens to study the alignment between states and their people. This paper presents a network-driven approach to analyze communities as they are established through different forms of bottom-up (e.g. Twitter) and top-down (e.g. UNGA voting records and international arms trade records) IR. By constructing and comparing different network communities we were able to evaluate the similarities between state-driven and citizen-driven networks. In order to validate our approach we identified communities in UNGA voting records during and after the Cold War. Our approach showed that the similarity between UNGA communities during and after the Cold War was 0.55 and 0.81 respectively (in a 0-1 scale). To explore the state- versus citizen-driven interactions we focused on the recent events within Syria within Twitter over a sample period of one month. The analysis of these data show a clear misalignment (0.25) between citizen-formed international networks and the ones established by the Syrian government (e.g. through its UNGA voting patterns).


Non-Symposium Articles


Motives for active participation in political blogs: A qualitative and quantitative analysis of eight German blogs / Kathrin Greuling & Thomas Kilian

Abstract: Blogs play an important role in political communication and in forming the opinions of political actors and politically interested users. Their ability to facilitate active interaction, present opinions, mobilize voters, and thereby, influence public opinion and eventually election outcomes makes political blogs an interesting research subject. Thus, this paper draws upon a qualitative content analysis to reveal the readers’ motivations and main themes to post comments and get involved in the discussions on political blogs. We analyze twelve blog discussions (500 pages) in eight left-leaning German political blogs and use qualitative content analysis to develop a category scheme comprising of 3868 codings. We find that social interaction with other blog users is the main motive for blogging on German political blogs together with the desire for the exchange of further information and analysis, and the expression of the own opinion concerning the subject. Moreover, we classify 191 users by their motivations for blogging using cluster analysis and identify the bloggers’ different communication roles. By providing insight into the users’ political discussions, motives for participation, and communication roles, we clarify the role that political blogs play in the process of forming political opinion and derive implications for politics.


Comparison of Smartphone and Online Computer Survey Administration / Tom Wells, Justin T. Bailey, & Michael W. Link

Abstract: The dramatic rise of smartphones has profound implications for survey research. Namely, can smartphones become a viable and comparable device for self-administered surveys? The current study is based on approximately 1,500 online U.S. panelists whom were smartphone users, and whom were randomly assigned to the mobile app or online computer mode of a survey. Within the survey, we embedded several experiments that had been previously tested in other modes (mail, PC web, mobile web). First, we test whether responses in the mobile app survey are sensitive to particular experimental manipulations as they are in other modes. Second, we test whether responses collected in the mobile app survey are similar to those collected in the online computer survey. Our mobile survey experiments show that mobile survey responses are sensitive to the presentation of frequency scales and the size of open-ended text boxes, as are responses in other survey modes. Examining responses across modes, we find very limited evidence for mode effects between mobile app and PC web survey administrations. This may open the possibility for multi-mode (mobile and online computer) surveys, assuming that certain survey design recommendations for mobile surveys are used consistently in both modes.


Response Quality of Self-Administered Questionnaires: A Comparison Between Paper and Web Questionnaires  / Vidal Díaz de Rada Igúzquiza & Juan Antonio Domínguez Álvarez

Abstract: The objective of this paper is to analyse the quality of the information collected by a self-administered survey addressed to citizens of Andalusia, who were offered the possibility of answering using the post or Internet. The study showed the advantages of using Web-based self-administered questionnaires. Web surveys showed a low number of unanswered questions, more detailed answers to open questions, and longer answers to questions than those generated from paper questionnaires. The five open questions with text showed longer answers from the Web survey, around 63 characters more. In the open questions with numerical answers, the use of drop-box (select list) generated better response than the use of a blank space left in the paper questionnaire.