Political predispositions, not popularity: people’s propensity to interact with political content on Facebook

ABSTRACT

Social media users are not just potential consumers of political content they are also potential producers and distributors. In this paper, we test whether political predispositions or the popularity of posts best explains users’ engagement with political content on Facebook. Using a large-scale survey deployed in Denmark, we utilize a 2 × 2 × 3 survey experiment that manipulates the partisan sponsor of a political message, the number of likes attributed to that message, and the nature of the comments attached to that post. Our findings indicate that individuals are most likely to like, comment, and share political content that aligns with their political predispositions, as the choice to like, share and comment political content on Facebook is largely unaffected by likes and comments from other users. Though we recognize the dangers of obstinacy in democratic discourse, we are somewhat assured by these findings, as it shows that those who engage with political content do not follow a blind herd mentality.

KEYWORDS:

• Political news
• social media
• political attitudes
• experiments

Disclosure statement

No potential conflict of interest was reported by the author(s).

Notes

1 The survey was in the field 23 March to 11 April 2017. The data collection was done in full accordance with all relevant regulations, including regulations at [blinded] and legal regulations in Denmark. Under these regulations, ethical approval for survey studies is not required. All participants had given informed consent prior to their participation in the study.

2 While the Social Democrats have historically had a less restrictive policy on immigrations, they had started to move rightwards on this issue some years prior to our experiment. By the time of our experiment their positions on immigration were therefore relatively similar to the policies of the Danish center-right parties Kosiara-Pedersen 2020; Stubager et al. 2021).

3 The exact number of likes chosen for our “many likes” treatments were based on the number of likes, which posts from these two politicians would typically end up with at the time of the experiment. The number of likes for our “few likes” treatment mimics the number of likes a post would have received almost immediately after being posted. In the period since our experiment, the number of users responding to Facebook posts from these two politicians has increased markedly, and their posts now typically receive far more likes and reactions.

Additional information

Funding

This work was supported by the Carlsberg Foundation under [grant number CF-15-0228].