Are Facebook likes polarizing society?

• Grant number: 21/06967-7
• Support Opportunities: Regular Research Grants
• Start date: January 01, 2022
• End date: December 31, 2023
• Field of knowledge: Applied Social Sciences - Communications
• Agreement: MCTI/MC
• Principal Investigator: Pablo Ortellado
• Grantee: Pablo Ortellado
• Host Institution: Escola de Artes, Ciências e Humanidades (EACH). Universidade de São Paulo (USP). São Paulo , SP, Brazil
• Associated researchers: Letícia Maria Costa da Nóbrega Cesarino ; Marcio Moretto Ribeiro
• Associated scholarship(s): 22/12370-6 - Are Facebook likes polarizing society?, BP.TT; 22/11886-9 - Are Facebook likes polarizing society?, BP.TT; 22/10957-0 - Are Facebook likes polarizing society?, BP.TT; + associated scholarships 22/11230-6 - Are Facebook likes polarizing society?, BP.TT; 22/11870-5 - Are Facebook likes polarizing society?, BP.TT; 22/11871-1 - Are Facebook likes polarizing society?, BP.TT; 22/11872-8 - Are Facebook likes polarizing society?, BP.TT; 22/11874-0 - Are Facebook likes polarizing society?, BP.TT - associated scholarships

Abstract

On Facebook and other social media messages cannot easily be targeted to different audiences of friends, co-workers and family, causing what the literature calls context collapse. Our hypothesis is that in this scenario where the specific context is lost, users can try to overcome this confusion by opting for the predominance of one of the contexts. Although some users develop strategies to preserve the segmentation of identities, adopting multiple accounts with pseudonyms, for example, it may be that other users end up adopting a dominant context for the messages. Likes received by different types of publications can function as quantitative clues that guide the choice of which identity should be imposed in the search for a univocal context. The audience for political posts may have offered more likes than audiences for other types of posts because they were organized and imbued with a militant will to reinforce the posts. To test this hypothesis, we intend to select a group of subjects with strong political identities and, with their permission, retrieve their post and likes history on Facebook. Each of these posts will be classified according to the context for which they appear to have been produced. This classification will be supported by life trajectory interviews. The objective is to check whether there is, over time, an increase in the number of political posts and whether this increase correlates with an increase in the number of political likes. Among these political posts, we will verify whether affirmations or rejections of identity or mere expressions of political opinion predominate. We also intend to analyze the pattern of likes of the subject in the posts of his contacts: that is, once the political context has imposed itself, verify whether the subject contributed to the dominance of this context among his contacts, in a process of viral diffusion of political polarization. (AU)