To date, the role of user behavior in the formation of politically homogeneous online environments (oftentimes called echo chambers) is not fully understood. Building on selective exposure research, we introduce the notion of selective political friending, that is, the preference for political like-mindedness in social affiliations on social networking sites. In a pre-registered laboratory experiment with users of social networking sites in Germany (N = 199), we find that users preferably build connections with those who share their opinions toward controversial (...) political issues. Political like-mindedness outperforms other friending criteria such as popularity or career-related fit with another user. Political friending is pronounced when individuals’ pre-existing opinions are strong. The present study points to the necessity to take the motivational complexity into account when studying phenomena linked to political homogeneity on SNS. (shrink)

Extremists often aim to paint a biased picture of the world. Radical narratives, for instance, in forms of internet memes or posts, could thus potentially trigger cognitive biases in their users. These cognitive biases, in turn, might shape the users’ formation of extremist attitudes. To test this association, an online experiment (N=392) was conducted with three types of right-wing radical narratives (elite-critique, ingroup-outgroup, violence) in contrast to two control conditions (nonpolitical and neutral political control condition). We then measured the impact (...) of these narratives on the activation of three cognitive biases of relevance in the formation of extremist attitudes: the ingroup-outgroup bias, the negativity bias, and the just-world hypothesis. The results indicate that violence narratives seem to be particularly harmful as they heighten participants’ negativity bias and increase just-world views. Just-world views in turn show a positive relationship to extremist attitudes, which highlights the need of regulating violence invocations on social media. (shrink)

Controversies in food and agriculture abound, with many portrayed as conflicts between polarized viewpoints. Framing such controversies as dichotomies, however, can at times obscure what might be a plurality of views and potential common ground on the subject. We used Q methodology to explore stakeholders’ views about pesticide safety, agricultural worker exposure, and human health concerns in the tree fruit industry of central Washington State. Using a purposive sample of English and Spanish-speaking agricultural workers, industry representatives, state agencies, educators, and (...) advocates, participants sorted 45 statements on pesticide use and perceived human safety risks in the tree fruit industry in 2011. We used PQMethod 2.33 statistical software program to identify viewpoints, based on differences between how participants sorted the statements. The results revealed three distinct viewpoints among 38 sorters that explained 52 percent of the variance. The viewpoints included the: skeptics who expressed concern over the environmental and human health impacts of pesticide use; acceptors who acknowledged inherent risks for using pesticides but saw the risks as known, small and manageable; and incrementalists who prioritized opportunities to introduce human capital and technological improvements to increase agricultural worker safety. We then brought representatives with these different viewpoints together to analyze the results of the Q study, and to brainstorm mutually acceptable improvements to health and safety in tree fruit orchards. In describing and analyzing this case study, we argue that Q methodology can serve as one potentially effective tool for collaborative work, in this case facilitating a process of orchard safety improvements despite perceived stakeholder polarization. (shrink)

(2013). Is Ideological Coverage On Cable Television An Ethical Journalistic Practice? An Examination of Duty, Responsibility, and Consequence. Journal of Mass Media Ethics: Vol. 28, No. 1, pp. 1-14. doi: 10.1080/08900523.2012.746533.