Twitter Debate Research Articles

Hallucination or vision? Establishing reliability and validity of AI-based qualitative analysis of the unified model of activism in the Scottish independence Twitter debate

PurposeThe present study explores both the validation of the unified model of activism and the methodological reliability of the LlamaParsing approach to natural language processing. Theoretically, it applies the unified model of activism within the context of the Scottish independence movement, evaluating its effectiveness in social media environment.Design/methodology/approachMethodologically, it addresses the reliability and validity challenges associated with AI analyses, particularly the issue of AI hallucinations—instances where AI generates seemingly accurate but incorrect information. By employing the LlamaParsing approach and then comparing and contrasting it with a quantitative content coding process, the study demonstrates how context-specific instructions can enhance the accuracy of AI analyses.FindingsThe findings indicate this approach not only tests and extends the unified model of activism but also offers a robust methodological framework for using NLP and RAG in qualitative research. This dual focus underscores the potential of AI to provide systematic and theoretically valuable insights while highlighting the importance of mitigating its limitations.Originality/valueThis study represents a cutting-edge approach to qualitative data analysis, theory development, and theory testing in communication using a tool that was developed in 2024.

The politics of platform folklore: Emotion, identity, and sense-making in far-right populist Twitter communities

Platforms are increasingly part of everyday life, but they remain opaque and impenetrable spaces for most users. To manage life on platforms, users thus need to engage with sense-making practices that help them understand and navigate online spaces. This paper studies how far-right populist activists interpret and navigate their presence on Twitter, using the concept of platform folklore, that is, unofficial and collective narratives aimed at relieving feelings of uncertainty associated with the opacity of platforms. The data consists of 20 life-history interviews with Swedish and American Twitter activists from right-wing populist communities, who all participated in populist Twitter debates. The analysis shows how platform folklore is constructed not only based on observations of content moderation, as emphasized in previous research, but also in correspondence with the political ideology and identities of the activists. This led them to interpret Twitter as a leftist organization that disfavored them, which stopped them from developing strategies to reach the goals they held as central to their Twitter activism. The paper concludes by discussing the role of emotions and ideology in platform folklore within political communities and suggesting directions for future research.

Entropy-based detection of Twitter echo chambers.

Echo chambers, i.e. clusters of users exposed to news and opinions in line with their previous beliefs, were observed in many online debates on social platforms. We propose a completely unbiased entropy-based method for detecting echo chambers. The method is completely agnostic to the nature of the data. In the Italian Twitter debate about the Covid-19 vaccination, we find a limited presence of users in echo chambers (about 0.35% of all users). Nevertheless, their impact on the formation of a common discourse is strong, as users in echo chambers are responsible for nearly a third of the retweets in the original dataset. Moreover, in the case study observed, echo chambers appear to be a receptacle for disinformative content.

The Challenges of Replicating Volatile Platform-Data Studies: Replicating Schatto-Eckrodt et al. (2020)

Replication studies in computational communication science (CCS) play a vital role in upholding research validity, ensuring reliability, and promoting transparency. However, conducting such studies in CCS often proves challenging due to the data environments’ dynamic nature and the complexities surrounding data and software sharing. To shed light on these challenges, we examine the replication process with CCS studies by computationally reproducing and replicating Schatto-Eckrodt et al.’s (2020) computational analysis of the X (formerly Twitter) debate about the term “gaming disorder” being added to the International Classification of Diseases 11. Our results indicate a reproduction success rate of 88.46% of the original findings. Replicating the analysis presents several obstacles, particularly in data access and availability. Five years after the original data collection, we were able to recollect only 55.08% of the initial sample, primarily due to user and platform activities, including account deletions, user suspensions, and privacy settings. Our reproduction and replication efforts revealed intricate challenges in conducting CCS research, particularly concerning data access and third-party platforms. To enhance replication in CCS, we emphasize the crucial role of data sharing, increased transparency, extensive documentation, and regulatory processes. Thus, our analysis underscores replications’ critical role in enhancing CCS research validity and reliability.

Pragmatic patterns and discourses on Twitter: Unpacking perspectives in the discussion of the Turów lignite mine

Much public debate today is carried out on Twitter (now X), where the participants employ a range of diverse resources to convey their ideas effectively. Disentangling the resources into clear and understandable structures and patterns presents a fresh challenge for discourse pragmatics. This article addresses that challenge methodologically by evaluating existing methods for identifying and classifying pragmatic patterns, revealing their drawbacks, and advocating for a new coding system. This categorizes tweets based on a post's primary pragmatic function (informing, appealing, or expressing emotivity), considering subsidiary functions. The study then applies this new scheme to analyze the Twitter debate on the Polish Turów lignite mine, which became a subject of international dispute as the Czech and European authorities sought the mine's closure to eliminate its negative environmental impact. The debate unfolds mainly in Polish, Czech, and English, each language being associated with a distinct discourse. The English discourse emphasizes a transnational appeal for widespread decarbonization, contrasting with the predominantly oppositional, national political perspectives in the Polish and Czech discourses. These rely heavily on emotivity directed, in the Polish case, primarily against the country's ruling party, and in the Czech one against the deal eventually reached with Poland to mitigate the problems.

Analyzing the Effects of Twitter on Electronic Media A Study of By-Elections 2022 in Pakistan

This research examines the impact of Twitter on electronic media in Pakistan during the by-elections of 2022. The study specifically focuses on the strategies utilized by political personalities, such as Imran Khan and Maryam Nawaz, to shape public conversation. Twitter, a highly influential social media tool, exerts significant influence in shaping public opinion and setting the agenda. This study uses agenda-setting theory to examine the impact of Twitter debates during the by-elections on the public's perception of prominent themes. This study explores eleven main themes, which include political issues and socio-economic factors. These themes contain topics such as "Lifafa Journalism," "Imported Government," "Lawlessness," and "Inflation." This study conducts a thorough examination of Twitter engagement, uncovering discrepancies in the frequency of tweets and accompanying media coverage received by Imran Khan and Maryam Nawaz on four prominent news networks. It is worth mentioning that Maryam Nawaz's tweets have a notable prevalence, particularly about topics such as "Harsh Language" and "Incompetent Government." On the other hand, Imran Khan's influence is evident in thematic domains such as "Imported Government," "Inflation," "Regime Change," and "Haqeeqi Azadi." The analysis of media coverage reveals that Imran Khan's tweets garnered significant visibility across many thematic categories, particularly on prominent news channels like as ARY News, Geo News, Sama News, and Dunya News. In contrast, the tweets of Maryam Nawaz received considerable attention, namely from Geo News and Dunya News, illustrating the varied media strategies utilized by each leader.

It’s Time to See the Forest for the Trees – Rhetorical Argumentation about Forest Use on Twitter

Twitter used to be a hotbed of topical societal debates. In this paper, we analyze discussions about Finnish forest use and conservation on Twitter during a five-month period in 2020–2021. By looking at the factual contents and the ways claims are justified, we ask how tweeters construct their arguments. The materials were analyzed by paying attention to background categories, content topic and style, and any additional information brought up by participants. Our findings show that a polarization between economic and ecological worldviews can be seen in Twitter debates, and that the participants use various methods to justify their stances. As a result, we also identified ways in which ordinary people take part in this debate by commenting, reframing, retweeting and sending their own photos as evidence.

Speaker landscapes: machine learning opens a window on the everyday language of opinion

ABSTRACT We propose a new method that embeds speakers into a spatial representation according to the linguistic similarity of their contributions to a debate. Such “speaker landscapes” can be constructed quantitatively using word embeddings by annotating text corpora of speech samples by tokens representing the speakers. The way embeddings are constructed from predictive machine learning models means that speaker-tokens are placed closer together if they are easier to confuse given their speech samples. The result is a nuanced measure of similarity in speech which takes into account the wealth of linguistic signification and structure that the text prediction model can learn. We validate this tool using two South African case studies: the Twitter debate around the arrest and imprisonment of former president Jacob Zuma, and quotes from the news media about land reform. The results show that speaker landscapes make social qualities such as group membership and opinions evident, and we discuss how speaker landscapes open up new methods for studying opinion discourse with text data.

Who’s fuelling Twitter disinformation on the COVID-19 vaccination campaign? Evidence from a computational analysis of the green pass debate

ABSTRACT The COVID-19 health emergency increased disinformation’s role and fostered a growing fragmentation between conflicting opinions on COVID-19 causes, vaccination policies, and government measures to deal with the pandemic. Studies have found that disinformation sources included private citizens, independent organizations, mainstream online newspapers and even public figures such as politicians, commentators, bloggers etc. In Italy, the Twitter debate ignited a conflict between mainstream positions in favour of restrictions, and more libertarian opinions extremely critical of government measures. Our research investigates, through a computational approach based on digital methods and social network analysis (SNA), opinion leaders’ roles in the Italian green pass debate on Twitter that surfaced in the second half of 2021. Drawing on the classic two-step model of communication, our essay identifies the Italian opinion leaders on Twitter and their content dissemination strategies. Our analysis reveals a limited number of dominant voices interacting in segregated networks of users. These networks can be considered echo chambers given the verbose and self-referential tweeting activity of their opinion leaders. Moreover, such activity involves spreading disinformation and conspiracy theories through a dissemination strategy aimed at diverting the audience from Twitter, towards ‘below-the-radar’ environments (e.g. Rumble), where political views are more radical.

“Will the law not protect survivors who don’t weep”: Twitter as a platform of feminist deliberation and democracy in India

An analysis of 40,000 tweets that trended after the Tarun Tejpal acquittal in India showed that the nature of the debate around issues of molestation and rape exhibited attributes of deliberation and demonstrated that Twitter in India, in certain cases, has strong potential to emerge as a space for deliberative feminist activism. Discussions gave impetus to advocacy around sexual molestation. While the word “victim” was used in more instances rather than the human rights–based term “survivor,” Twitter debates were supportive toward survivors of assault. There was minimum trolling and patriarchy was called out as was a legal system that sided with the influential man of power. Although city-bred English-speaking voices dominated, conversations were intersectional in nature acknowledging how the horror of physical assault was perceived by different women belonging to disparate socio-economic strata and how legal systems exacerbated gender related crimes.

Bow-tie structures of twitter discursive communities

Bow-tie structures were introduced to describe the World Wide Web (WWW): in the direct network in which the nodes are the websites and the edges are the hyperlinks connecting them, the greatest number of nodes takes part to a bow-tie, i.e. a Weakly Connected Component (WCC) composed of 3 main sectors: IN, OUT and SCC. SCC is the main Strongly Connected Component of WCC, i.e. the greatest subgraph in which each node is reachable by any other one. The IN and OUT sectors are the set of nodes not included in SCC that, respectively, can access and are accessible to nodes in SCC. In the WWW, the greatest part of the websites can be found in the SCC, while the search engines belong to IN and the authorities, as Wikipedia, are in OUT. In the analysis of Twitter debate, the recent literature focused on discursive communities, i.e. clusters of accounts interacting among themselves via retweets. In the present work, we studied discursive communities in 8 different thematic Twitter datasets in various languages. Surprisingly, we observed that almost all discursive communities therein display a bow-tie structure during political or societal debates. Instead, they are absent when the argument of the discussion is different as sport events, as in the case of Euro2020 Turkish and Italian datasets. We furthermore analysed the quality of the content created in the various sectors of the different discursive communities, using the domain annotation from the fact-checking website Newsguard: we observe that, when the discursive community is affected by m/disinformation, the content with the lowest quality is the one produced and shared in SCC and, in particular, a strong incidence of low- or non-reputable messages is present in the flow of retweets between the SCC and the OUT sectors. In this sense, in discursive communities affected by m/disinformation, the greatest part of the accounts has access to a great variety of contents, but whose quality is, in general, quite low; such a situation perfectly describes the phenomenon of infodemic, i.e. the access to “an excessive amount of information about a problem, which makes it difficult to identify a solution”, according to WHO.

The architecture of partisan debates: The online controversy on the no-deal Brexit.

We propose a framework to analyse partisan debates that involves extracting, classifying and exploring the latent argumentation structure and dynamics of online societal controversies. In this paper, the focus is placed on causal arguments, and the proposed framework is applied to the Twitter debate on the consequences of a hard Brexit scenario. Regular expressions based on causative verbs, structural topic modelling, and dynamic time warping techniques were used to identify partisan faction arguments, as well as their relations, and to infer agenda-setting dynamics. The results highlight that the arguments employed by partisan factions are mostly constructed around constellations of effect-classes based on polarised verb groups. These constellations show that the no-deal debate hinges on structurally balanced building blocks. Brexiteers focus more on arguments related to greenfield trading opportunities and increased autonomy, whereas Remainers argue more about what a no-deal Brexit could destroy, focusing on hard border issues, social tensions in Ireland and Scotland and other economy- and healthcare-related problems. More notably, inferred debate leadership dynamics show that, despite their different usage of terms and arguments, the two factions' argumentation dynamics are strongly intertwined. Moreover, the identified periods in which agenda-setting roles change are linked to major events, such as extensions, elections and the Yellowhammer plan leak, and to new issues that emerged in relation to these events.

Measuring user engagement with low credibility media sources in a controversial online debate

We quantify social media user engagement with low-credibility online news media sources using a simple and intuitive methodology, that we showcase with an empirical case study of the Twitter debate on immigration in Italy. By assigning the Twitter users an Untrustworthiness (U) score based on how frequently they engage with unreliable media outlets and cross-checking it with a qualitative political annotation of the communities, we show that such information consumption is not equally distributed across the Twitter users. Indeed, we identify clusters characterised by a very high presence of accounts that frequently share content from less reliable news sources. The users with high U are more keen to interact with bot-like accounts that tend to inject more unreliable content into the network and to retweet that content. Thus, our methodology applied to this real-world network provides evidence, in an easy and straightforward way, that there is strong interplay between accounts that display higher bot-like activity and users more focused on news from unreliable sources and that this influences the diffusion of this information across the network.

Sephora’s Starter Witch Kit

In late summer 2018, beauty chain Sephora announced the release of a “Starter Witch Kit” in collaboration with fragrance company Pinrose. By September, Sephora announced it was cancelling the product after receiving extensive criticism on social media, particularly from Modern Witches. This article examines the uproar surrounding Sephora’s Starter Witch Kit as it played out on Twitter. The debate on Twitter included Witches protesting the appropriation and commodification of their sacred traditions, as well as outsiders who questioned the right of Witches to complain about spiritual theft. This Twitter debate was an opportunity for Modern Witches to substantiate and legitimize their identities as Witches. Witches distinguished their identities as “authentic” by mocking certain products and consumers, and demarcated practices/traditions as distinctive of Witchcraft by calling them sacred. By accusing Sephora of spiritual theft, Witches also largely elided their own engagement with appropriation from religious traditions.

The Expression of Moral Values in the Twitter Debate: a Corpus of Conversations

The present work introduces MoralConvITA, the first Italian corpus of conversations on Twitter about immigration whose annotation is focused on how moral beliefs shape users interactions.The corpus currently consists of a set of 1,724 tweets organized in adjacency pairs and annotated by referring to a pluralistic social psychology theory about moral values, i.e. the Moral Foundations Theory (MFT). According to the MFT, different configurations of moral values determines the stance of individuals on sensitive topics, such as immigration, civil rights, and gender equality. Moreover, an annotation of adjacency pairs’ conversational dynamics is provided.Results of the analysis we applied on MoralConvITA shows that interesting patterns occur in the corpus, emerging from the intersection of moral studies and pragmatics that need to be generalized over larger corpora, and shedding some light on a novel promising perspective on the inter-user dynamics occurring in social media.

Migrant Racialization on Twitter during a border and a pandemic crisis.

This work examines how the outbreak of the COVID-19 pandemic reshaped the migration debate on Twitter. Through co-hashtag network analysis, time-frequency and content analysis, it shows that the pandemic was related with positive (humanitarian) and negative (threat) stances about migration. The positive side focused on the need to protect refugees stranded at camps in Greece from COVID-19. The negative focused on the Greek-Turkish land-border crisis (Evros crisis), using COVID-19 to reinforce migrants as racialized others. These findings fit the problematization of positive and negative migrant representations in the Global north as Eurocentric. In the case of camps, refugees fit well within the victim/helpless frame, justifying humanitarianism, this time on health grounds. Regarding the border crisis, refugees also fit the Eurocentric frame of violent/male/inferior other who could spread a deadly virus. Overall, COVID-19 intertwined with migration in Twitter debates, reinforcing the racialized, Eurocentric representational field on migrants from the Global south.

Online Performance of Civic Participation: What Bot-like Activity in the Persian Language Twittersphere Reveals About Political Manipulation Mechanisms

In this paper, we reveal the understudied transnational dimensions of politically manipulative activity on social media. Specifically, we identify and investigate a bot-like Twitter network associated with the controversial organization of Iranian political exiles, the Mojaheddin-e Khalgh (MEK). Tracing and contextualizing the Twitter debate around women’s rights within the 2016 Iranian Parliamentary election, our analysis contributes to the scholarship on diaspora and digital media by drawing attention to the often-neglected potentials for non-state actors such as diaspora groups to make use of social media to promote political propaganda that advances militarist violence. We demonstrate how the MEK network’s “online performance of civic participation” is typical of a bot-net of weak influence inside Iran, but that the aims and extent of its influence can only be fully understood by situating it within a historical and transnational analysis of Iranian diasporic media and politics, one that takes complex US-Iran diplomacy dynamics into consideration.

Geographies of Twitter debates

Over the last years, the prodigious success of online social media sites has marked a shift in the way people connect and share information. Coincident with this trend is the proliferation of location-aware devices and the consequent emergence of user-generated geospatial data. From a social scientific perspective, these location data are of incredible value as it can be mined to provide researchers with useful information about activities and opinions across time and space. However, the utilization of geo-located data is a challenging task, both in terms of data management and in terms of knowledge production, which requires a holistic approach. In this paper, we implement an integrated knowledge discovery in cyberspace framework for retrieving, processing and interpreting Twitter geolocated data for the discovery and classification of the latent opinion in user-generated debates on the internet. Text mining techniques, supervised machine learning algorithms and a cluster spatial detection technique are the building blocks of our research framework. As real-word example, we focus on Twitter conversations about Brexit, posted on Uk during the 13 months before the Brexit day. The experimental results, based on various analysis of Brexit-related tweets, demonstrate that different spatial patterns can be identified, clearly distinguishing pro- and anti-Brexit enclaves and delineating interesting Brexit geographies.

Why it is important to consider negative ties when studying polarized debates: A signed network analysis of a Dutch cultural controversy on Twitter.

Despite the prevalence of disagreement between users on social media platforms, studies of online debates typically only look at positive online interactions, represented as networks with positive ties. In this paper, we hypothesize that the systematic neglect of conflict that these network analyses induce leads to misleading results on polarized debates. We introduce an approach to bring in negative user-to-user interaction, by analyzing online debates using signed networks with positive and negative ties. We apply this approach to the Dutch Twitter debate on 'Black Pete'-an annual Dutch celebration with racist characteristics. Using a dataset of 430,000 tweets, we apply natural language processing and machine learning to identify: (i) users' stance in the debate; and (ii) whether the interaction between users is positive (supportive) or negative (antagonistic). Comparing the resulting signed network with its unsigned counterpart, the retweet network, we find that traditional unsigned approaches distort debates by conflating conflict with indifference, and that the inclusion of negative ties changes and enriches our understanding of coalitions and division within the debate. Our analysis reveals that some groups are attacking each other, while others rather seem to be located in fragmented Twitter spaces. Our approach identifies new network positions of individuals that correspond to roles in the debate, such as leaders and scapegoats. These findings show that representing the polarity of user interactions as signs of ties in networks substantively changes the conclusions drawn from polarized social media activity, which has important implications for various fields studying online debates using network analysis.

The missing link between overtourism and post-pandemic tourism. Framing Twitter debate on the Italian tourism crisis

PurposeThis paper aims to exploit existing tourism knowledge to frame the unprecedented pandemic tourism crisis, its key aspects and impacts on the tourism industry. It builds a conceptual bridge and discusses the opportunity to capitalise on the missing link between the pre-COVID overtourism and the post-COVID “undertourism” debates.Design/methodology/approachA cross-fertilisation between the overtourism knowledge and the emerging COVID-19 literature stream is proposed and supported by an online media analysis focussing on the Italian tourism debate on Twitter. A text analysis of 2,500 posts helps discuss the conceptual framework.FindingsThe analysed Twitter debate prioritised socio-economic impacts, regulative actions and the recovery approach, representing government as the pivotal actor to overcome the pandemic crisis. An integrative interpretative framework results from this research, opening three areas of inquiry, such as the recovery–reform continuum, managerial approaches beyond regulative frames of action and a critical sizing of digital technologies deployment.Research limitations/implicationsSamples with different geographical and temporal coverage may provide further and multifaceted insights into the emerging tourism online media debate.Originality/valueAn original conceptualisation counter-intuitively frames post-pandemic tourism scenarios. Additional elements of originality are the online media analysis contributing to the emerging COVID-19 agenda and the use of Twitter social platform to investigate the tourism debate.