Saif ShahinTilburg University | UVT · Department of Culture Studies
Saif Shahin
PhD, University of Texas at Austin
About
45
Publications
33,317
Reads
How we measure 'reads'
A 'read' is counted each time someone views a publication summary (such as the title, abstract, and list of authors), clicks on a figure, or views or downloads the full-text. Learn more
509
Citations
Introduction
My research interests include critical data studies, digital culture, and media and politics in global contexts. I am especially interested in data and technology as sociocultural phenomena, power in online social networks, and the politics of online identity construction. I work with qualitative, quantitative, and computational research methods—including machine learning and network analysis.
Additional affiliations
August 2018 - present
August 2016 - July 2018
Education
January 2012 - June 2017
Publications
Publications (45)
India and China have launched enormous projects aimed at collecting vital personal information regarding their billion-plus populations — and building the world’s biggest datasets in the process. However, both Aadhaar in India and the Social Credit System in China are controversial and raise a plethora of political and ethical concerns. The governm...
This article examines how the nationalist imagination structures cyberspace from the bottom up, or what I call “user-generated nationalism.” It also looks at the interplay between nationalism and other, non-spatial modes of social identification. My analysis of a month of tweets indicates that religious, racialized, and partisan identities are quit...
This study examines the global diffusion of Black Lives Matter as digitally networked connective action. Combining social network analysis with qualitative textual analysis, we show that BLM was hybridized in different ways to give voice to local struggles for social justice in Brazil, India, and Japan. However, BLM’s hybridization stirred right-wi...
This article adopts a poststructural approach to examine the relationship between the news media and international relations. It compares 15 years of international aid coverage from two donor nations, the United States and Britain, and two recipient nations, India and Pakistan, to understand the types of identities news media construct for a nation...
Protest movements around the world have become increasingly likely to incite counterprotests that adopt an opposing stance. This study examines how a protest and a counterprotest interact with and shape each other as digitally networked connective action. My empirical focus is the so-called Million MAGA March—in which supporters of U.S. President D...
This study examines how perceived differences in the affordances of social media platforms influence users' willingness to express opinions on a controversial issue, viz., systemic racism. Drawing on a U.S. nationally representative survey, our analysis suggests that fear of social isolation has a significant effect on Facebook but not on Twitter....
We theorize network diplomacy as a form of symbolic interaction and examine Twitter as an arena for such interaction by comparing the social networks of U.S. missions in Britain, India, and China. Network analysis of a year’s sample of tweets, followed by manual coding of “strong ties” in each country, reveals the three networks reflect and reprodu...
We theorize network diplomacy as a form of symbolic interaction and examine Twitter as an arena for such interaction by comparing the social networks of U.S. missions in Britain, India, and China. Network analysis of a year’s sample of tweets, followed by manual coding of “strong ties” in each country, reveals the three networks reflect and reprodu...
An emerging line of research has drawn attention to the significance of national identity in shaping digital diplomatic practices. In this study, we look at the reciprocal construction of national identity on Twitter by corresponding foreign missions. Specifically, we examine one year of Twitter posts from the South Korean missions in Japan and the...
This research examines how protesters supporting Donald Trump and counterprotesters opposed to his re-election as US president interpreted and reinterpreted “flags” as political symbols. The empirical analysis involves the computational modeling of two corpora of Twitter posts using the hashtags of protesters and counterprotesters, respectively. Th...
The fake news crisis has generated a veritable cottage industry of books and academic articles over the past few years. And yet, so much more needs to be said. Two new books take the conversation forward with varying degrees of success. Justin P. McBrayer’s monograph offers a thought-provoking exposition of not just the supply of but also the deman...
Facial recognition is one of the most contentious applications of artificial intelligence. In 2019, the US state of California passed The Body Camera Accountability Act (AB-1215), banning police from using facial recognition technology on body cameras for three years. This article traces the trajectory of AB-1215 as a social discourse from its firs...
Twenty years after the 9/11 attacks, stereotypes that associate Muslims with terrorism go far beyond depictions in newspapers and television. Recent research raises the alarm about rampant Islamophobia in digital spaces, particularly far-right groups’ use of disinformation and other manipulation tactics to vilify Muslims and their faith.
A substantial body of scholarship, emerging over decades, links U.S. media coverage of international affairs to foreign policy objectives. This is disturbing because it not only brings into question the supposed independence of the media as an institution—a basic normative expectation in a democracy—but also because people depend primarily upon the...
Mobile money transfer technology has diffused widely in recent years, especially in the Global South. Our study adopts a constructivist approach to understanding mobile money as a ‘technosocial’ artifact, which is shaped by social forces even as it reshapes society over time. By analyzing local news discourses from Kenya, Bangladesh, and India betw...
Connective action, or individuals networking with each other online to form social movements, rarely leads to lasting change. In this study, we argue that such movements are ultimately ineffective because they struggle to sustain themselves over time, and identify the reasons behind their transience. Our analysis focuses on Twitter conversations ab...
Hiding behind anecdotes and statistics is a deeper truth about disinformation: its acceptance relies less upon the content of a campaign itself and more upon how closely it coheres with an individual’s beliefs about the world they live in—beliefs that are increasingly built around partisan boundaries. Discrete pieces of disinformation do not carry...
This study places the “cognitive elaboration model” on news gathering and political behavior within the dual-processing “elaboration likelihood model” to derive hypotheses about the effects of incidental news exposure and tests them using two-wave panel data. Results indicate incidental news exposure predicts online participation but not offline pa...
This article maps patterns of interest in key terms associated with copyright and online culture in the US context. Using exploratory factor analysis of data from Google Trends, authors examined patterns in keyword searches between 2004 and 2019. The data show three distinct periods of interest. The first period consists of utopian, cause-driven se...
This study advances the protest paradigm as a transnational theory by examining how ideological affiliations within and across national borders influence the framing of a protest movement. Our empirical focus is the coverage of the 2016-17 South Korean “candlelight” protests to oust conservative President Park Gyun-hye in Korean and U.S. newspapers...
Political campaigns mostly run parallel to each other during an election cycle, but intersect when the main candidates face off for televised debates. They offer supporters of these candidates a chance to engage with each other while being exposed to views and opinions different from their own. This study uses a combination of social network analys...
The alt-right White Nationalist movement, which emerged in response to the election of America’s first Black president, adopted Twitter from the outset. Tracing its evolution over eight years in retweets, our study suggests that the movement was relatively small and factionalized until 2015—but its subgroups closed ranks following Donald Trump’s ca...
Taqwacore is a musical genre that fuses punk rock with South Asian and Middle Eastern influences. It is also a post-9/11 cultural phenomenon involving mostly young American and European Muslim artists trying to make sense of a world falling apart in song and music. This article explores how young American-Muslims perform complex, contradictory iden...
Drawing on the constructivist tradition in international relations, we examine the influence of national identity-or how a nation views itself in relation to other nations-on the tweeting practices of its diplomatic missions. Our analysis focuses on the use of Twitter by U.S. missions in Britain, India, and China over a four-month period brimming w...
Social networking sites can help global aid agencies converse with the communities they work with as well as listen to public conversations on vital issues. This study develops a technosocial framework that specifies how different affordances of Twitter—from the topical content of tweets to replies, retweets, hashtags, and hyperlinks—relate to diff...
Two deadly explosions took place in the United States, two days apart from each other, in the middle of April 2013. The Boston marathon bombings of April 15 killed three people and shook the nation. The blast at a fertilizer plant on the outskirts of West, Texas on April 17 claimed 15 lives but hardly left a mark on the national consciousness. In t...
This study traces how Facebook-promoted internet.org/Free Basics, despite initial acclaim, was eventually rejected in India – and how net neutrality came to be codified in the process. Topic modeling of articles (N=1,752) published over two-and-a-half years in 100 media outlets pinpoints the critical junctures in time at which the public discourse...
The role of the media, and especially the social media, in the Arab Spring has been extensively debated in academia. This study presents a survey of studies published in scholarly journals on the subject since 2011. We find that the bulk of the research contends that social media enabled or facilitated the protests by providing voice to people in s...
Raw data collected through surveys, experiments, coding of textual artifacts or other quantitative means may not meet the assumptions upon which statistical analyses rely. The presence of univariate or multivariate outliers, skewness or kurtosis in a distribution, and heteroscedasticity or multicollinearity among variables may compromise data analy...
Big Data is having a huge impact on journalism and communication studies. At the same time, it has raised a plethora of social concerns ranging from mass surveillance to the legitimization of prejudices such as racism. This article develops an agenda for critical Big Data research. It discusses what the purpose of such research should be, what pitf...
Taqwacore is a musical genre that fuses punk rock with South Asian and Middle Eastern influences. It is also a cultural phenomenon involving mostly young Muslim artists living in the West, negotiating their hyphenated identities in song and music. Just as interesting as the phenomenon itself is the story of its birth out of a work of fiction, The T...
This study examines U.S. and British media coverage of the “right to be forgotten” in the light of their legal approaches and public attitudes toward privacy. Algorithmic and qualitative textual analysis techniques are combined to uncover the ideologies and interests that structure the discourse and shape its outcome. The analysis reveals that U.S....
This study assesses the scope and applicability of the “protest paradigm” in non-Western contexts by examining the news coverage of Brazilian, Chinese, and Indian protests in their domestic media. Two publications from each nation, one conservative and one progressive, are content analyzed for adherence to a series of marginalization devices that h...
As computer-assisted research of voluminous datasets becomes more pervasive, so does the criticism of its epistemological, methodological, and ethical/normative inadequacies. This article proposes a hybrid approach that combines the scale of computational methods with the depth of qualitative analysis. It uses simple natural language processing alg...
This article proposes a theoretical framework for understanding modernity as lying at the intersection of two dimensions: (1) the narrative of modernity as interpreted variously in particular nations and (2) the metanarrative of modernity as a universal goal that nations tend to share. It demonstrates that interpretations of modernity vary among na...
This study proposes a dichotomous set of frames, the Blame Frame and the Explain Frame,to examine how the news media cover sudden tragic events. The Blame Frame affixes responsibility on human agents and foregrounds the pursuit of punishment and justice. TheExplain Frame takes responsibility away from human agents and describes the tragedy in terms...
This study examines how two publications with a common religious affiliation—“Muslim/Islamic”—but different racial affiliations—“indigenous/Black” and “immigrant/Arab”—frame news events. It develops two interrelated ideas. First, identity is not simply an “individual level” but also a higher, “organizational level” of influence on news. Second, new...
The growing scholarly literature on Muslims and the media in the United States has paid little attention to the American-Muslim press. This study compares the coverage of two major American-Muslim publications, the bimonthly news magazine Islamic Horizons and the weekly tabloid Muslim Journal, at four key moments beginning with 9/11. Content analys...
Questions
Questions (4)
I am looking for a way to remove stopwords from a Spanish-language corpus of news articles as part of preprocessing for topic modeling. The package I use, topicmodels, only seems to have a dictionary of English-language stopwords.
A couple of reviewers have asked me to take out tweets that may have been posted by "Russian bots" from my dataset before the analysis. Such tweets supposedly undermine the results. This has led me to wonder:
1) Is there a way to do that? How would you even know that particular tweets are posted by bots (or specifically "Russian" bots)?
2) Bots, Russian or otherwise, posting on social media is not a new phenomenon. I've been to conferences where bot developers are hailed for automating the posting process. I haven't come across any social data analyses that take steps to remove posts by bots. So why is this suddenly a concern? And, supposing there are going to be ways to remove bot posts, does this new issue undermine the validity of previous research using social data that did not remove bot posts?
I used sentiment analysis and topic modeling to analyze ~4 million tweets for a study. A reviewer has asked me why I didn't analyze a random sample instead. Is there any reason why it would be better to study a sample rather than the entire corpus?
I am trying to do a topic modeling study of a dataset of about 4 million tweets using Mallet and running into issues with working memory, or "heap space." My computer does have around 15 GB of working memory, but Mallet, by default, utilizes only 1 GB. So I was getting the following error:
Exception in thread "main" java.lang.OutOfMemoryError: Java heap space
I expanded the Mallet heap space allocation in the manner prescribed on https://programminghistorian.org/lessons/topic-modeling-and-mallet#issues-with-big-data. But it didn't help. So I was wondering if anyone had a solution.
Thanks.
