Deen Freelon’s research while affiliated with University of North Carolina at Chapel Hill and other places

... The spread of false information on social media poses risks to public health [1], democratic processes [2], and social cohesion [3]. Social media has been broadly observed to preferentially support the spread of false news over true news [4,5,6,7]. Scholars as well as social media platforms are actively working to design and test strategies to limit its transmission [8,9,10], including fact-check warning labels placed on individual sources or pieces of information [11,12], educational interventions to boost users' competencies at identifying false information [13,14,15,16], and a shift to design objectives other than user engagement [17,18,19]. ...

Reference:

Community Notes Moderate Engagement With and Diffusion of False Information Online

... However, the chronological feed did not significantly alter levels of political polarization, knowledge, or participation. Allcott et al. (2024) conducted a large-scale randomized experiment involving over 35,000 users who were paid to deactivate their accounts for six weeks before the 2020 US election. The study found that deactivating Facebook and Instagram reduced political participation, mainly online, and had no significant effect on political knowledge overall. ...

Reference:

Computational communication research (CCR): An early and active adopter of computational social science (CSS) Special Issue

... Empirical research on data linkage focuses on online news exposure, political knowledge and participation (Guess et al., 2023;Kristensen et al., 2017;Song & Cho, 2021;Thorson et al., 2021), leveraging social media data from Facebook (Thorson et al., 2021), Twitter (Freelon et al., 2024), and Instagram (Driel et al., 2022). So far, there has been no research leveraging data linkage to understand selective engagement with health-related TikTok videos. ...

Reference:

TikTok Engagement Traces Over Time and Health Risky Behaviors: Combining Data Linkage and Computational Methods

... According to research, conspiracy beliefs might be extremely difficult to change straightforwardly: believers in conspiracy theories may overlook factual information, selectively pick certain pieces of information or assume that fact-checkers are part of the conspiracy theories themselves (Tingley & Wagner, 2017). In this regard, Freelon (2024) highlighted that conspiracy beliefs and prejudice could have a closer conceptual and normative relationship than previous studies have shown: they could manifest in the same people, produce similar detrimental effects against the outgroup and benefit from the same interventions. Hence, research started to underline the importance of finding alternative approaches to reduce conspiracy beliefs and one of the possible strategies can be identified in intergroup contact (Allport, 1954;Jolley et al., 2023). ...

Reference:

The more positive intergroup contacts you have the less LGBTQ conspiracies you will report: The role of knowledge, anxiety, and empathy

... Interestingly, age differences emerged in our study for sources of information; however, rather than differing across types of sources, our results indicated a difference in the amount of information received by a source: media. Media has recently shown to be a promising tool to deliver important information about sexual violence; for example, a study found that adolescent males (aged 15-19) who were exposed to short social media videos on sexual violence (via TikTok) had higher knowledge of consequences and perceived severity of sexual violence compared to controls (Nicolla et al., 2023). Our findings indicate that older youth access media as a source of information more than younger youth. ...

Reference:

Teen Dating Violence: Youths’ Knowledge and Experience of Unhealthy Relationships

... Although our experimental design allowed for tightly controlled testing conditions, future work should explore real social media environments using randomized controlled trials because more variables are involved such as the influence of content algorithms (Brady, Jackson, et al., 2023;Guess et al., 2023). Although intentions to share are correlated with actual sharing (Mosleh et al., 2020), such field experiments would also extend our work by This document is copyrighted by the American Psychological Association or one of its allied publishers. ...

Reference:

Social Identity Shapes Antecedents and Functional Outcomes of Moral Emotion Expression

... In dit kader wordt veelvuldig verwezen naar de Amerikaanse presidentsverkiezingen in 2016 waar vanuit Rusland door hackers en bots een grote hoeveelheid politieke disinformatie werd verspreid en gerichte campagne tegen de Democratische presidentskandidaat Hillary Clinton werd gevoerd (Bennett & Livingston, 2018). Ook in de presidentsverkiezingen van 2020 wordt veelvuldig gewezen op de aanwezigheid van desinformatie (Guess et al., 2023) en er is inderdaad bewijs dat buitenlandse entiteiten, en met name Rusland, actief proberen de verkiezingen te beïnvloeden. Zo laat een studie van Kim et al. (2018) Ook in analyses van de Brexit campagne wordt de verspreiding van desinformatie onderscheiden als een strategie om de uitkomst te beïnvloeden -alhoewel de aandacht daar vooral uitgaat naar verspreiding vanuit binnenlandse opinieleiders en ook traditionele media (Gaber en Fisher, 2022). ...

Reference:

Verkenning en verdieping democratische erosie in Nederland

... Third, Mechanical Turk samples and Twitter (X) users who post links to debunked articles are highly non-representative of the American public. For instance, the majority of users in our field experiment are politically rightleaning -which, while non-representative of the general public, is reflective of empirical work showing that political conservatives are more likely to share and be exposed to low-quality content online [45][46][47][48][49]. And fourth, our field experiment and survey experiment 1 were conducted within the year leading up to the 2020 U.S. Presidential election. ...

Reference:

Promoting engagement with social fact-checks online: Investigating the roles of social connection and shared partisanship

... It is also much easier to recruit representative participants because subjects can work remotely online. Since everything is virtual on digital platforms, the intervention stimuli could be much more natural (e.g. by altering the algorithms; see Kramer et al., 2014;Nyhan et al., 2023). Finally, unlike offline experiments, computational experiments could trace the user log data and measure the process and outcomes unobtrusively and computationally. ...

Reference:

Computational communication research (CCR): An early and active adopter of computational social science (CSS) Special Issue

... 8), have been slow to change. 3 Although there are plenty of benefits to subfields having identifiable parameters and shared understandings, our concern is that many scholars studying political communication through the lens of identity are excluded from agenda-setting spaces and key journals (see Freelon, Pruden, Eddy, et al., 2023a;Freelon, Pruden, & Malmer, 2023b) because they do not use quantitative methods and/or because they theorize identity in ways incorrectly interpreted as too niche for such spaces. Of course, this problem is not limited only to political communication. ...

Reference:

Dialogue on difference: Identity and political communication