Saturday, January 5, 2019

Individuals in a politically homogeneous social networking site environment were more likely to unfriend/hide the political rivals; liberals are more prone to unfriending, as are the educated; the young hide

Social networking site as a Political Filtering Machine: Predicting the Act of Political Unfriending and Hiding on Social Networking Sites. Joseph Yoo, Margaret Yee Man Ng, Thomas Johnson. The Journal of Social Media in Society, Vol 7, No 2 (2018),

Abstract: Social networking sites (SNSs) seem to have become a political filtering machine that allows users to classify their online friends based on their political ideologies. Hiding and unfriending on social media has turned into being a political gesture, discriminating individuals with opposite political views on SNSs. Unfriending activities during the 2016 U.S. Presidential election and during the Hong Kong Umbrella Movement are two notable examples. Individuals have the tendency to consume politically congenial information and be surrounded by people who share the same views. The formation of echo chambers and the reason for relationship dissolution on SNSs can be explained by Social Identity Theory.

Through an online panel survey of 386 SNS users, this study examined how factors of political ideologies, social media and offline political participation and likeminded exposure on SNSs can predict hiding and unfriending/unfollowing on Twitter and Facebook. Results from ordinary least square (OLS) regression analysis revealed that if individuals had been in a politically homogeneous SNS environment, they were more likely to unfriend, suggesting the reinforcement of echo chambers in SNSs. Both social media and offline political participations predicted the dissociative, indicating that unfriending and hiding could be regarded as a new form of online political participation to engage in political affairs.

Keywords: Political participation, Hiding, Unfriending, Relationship dissolution, like-minded exposure

No comments:

Post a Comment