The existence of the internet and social media can have a positive impact on democracy by providing autonomy for individuals in creating content and obtaining unlimited information that can be accessed by anyone (Dahlberg, 2007). However, the open channels on the internet also change how agenda setting and propaganda can be done by anyone. In the research of Shin & Thorson (2017) and Ardi (2019), partisans of internet users were caught several times not checking facts for misleading, provocative news with unaccountable truth. They selectively only share messages according to their tastes (Shin & Thorson, 2017). Garrett (2009) states that individuals tend to be motivated by the desire to justify their political opinions when using the media. This motivation then makes the individual have a confirmation bias.
Confirmation bias is seen as a tendency to support information that supports one’s initial assumptions and thoughts rather than exploring information in a more scientific and neutral manner (Hogg & Vaughan, 2011). The investigation of confirmation bias on social media is motivated by the echo space phenomenon where it turns out that the algorithms on social media have led users to strengthen the assumptions they already have. This phenomenon occurs because of priority in search engine results and newsfeeds in social media is only based on user activity, such as sharing history, clicks, likes, comments, and searches (Bessi, 2016; Jamieson & Cappella, 2008; Tufekci, 2015). This makes most users look like frogs in a shell when they see certain events that are only directed based on their taste (Montag, 2019). The internet has made the individual the focal point for information consumption and production. On the other hand, media algorithms have also made individuals who tend to be uncritical, closed-minded, perceive their group as superior to others, become trapped more in their biased beliefs.
For internet users, critical thinking has a fundamental role as an initial filter to decide whether to trust information before sharing it with a wider audience. Critical thinking is closely related to cognitive problems and is closely related to openness to experience (Clifford, Boufal, & Kurtz, 2004). In this research, critical thinking will be viewed in two independent dimensions, namely: 1) the ability to exercise reflective skepticism; and 2) critical openness. Someone who has two abilities in the critical thinking dimension is thought to tend to minimize their confirmation bias.
Other factors related to confirmation bias are authoritarian personality and perceptions of collective entitlement. Lavine et al. (2005) found that individuals who tend to have authoritarian personalities tend to selectively only choose information that is in line with the attitudes they have taken. Meanwhile, the concept of collective rights is assumed to be related to how individuals tend to think that their group is more appropriate and entitled to everything. Groups that tend to have this kind of perception also tend to experience confirmation bias by selecting information that supports their group. Furthermore, research also looks at how the relationship between political preferences (eg pro incumbent and pro-opposition presidential candidates) and confirmation bias is based on contextual phenomena.
This study explicitly wants to investigate whether there is a correlation between political preferences (manifested in the election of presidential candidates) and confirmation bias. It will also be seen how the role of each variable (ie critical openness, reflective skepticism, authoritarian personality; and the perception of collective rights) can predict the tendency for confirmation bias.
Method and results
The study was conducted in 2017 to students who were partisans in an extra-campus organization at a tertiary institution in Surabaya. The total participants were 95 people. Data mining was carried out by directly visiting partisans who were willing to become respondents in this study. Respondents were aged between 18 to 24 years, of which 59% were male and 41% were female. 44% of participants reported spending more than four hours online a day, 24% spending three to four hours a day, 19% spending two to three hours a day, while the rest spent less than two hours a day. Sixty-one percent reported religious nationalism as their preferred ideology, 14% preferred nationalism, 7% wanted a state based on religious law, 11% identified as democratic socialists. As many as 37% of students supported the incumbent presidential candidate and 46% supported the opposition candidate, while the rest abstained.
Several instruments were used for the purposes of this study. The instrument for measuring confirmation bias was purposely designed for this study. Other instruments are the adoption of instruments that have been validated: the critical thinking scale (consisting of two dimensions, reflective skepticism and critical openness), the authoritarian personality scale, and the scale of perceptions of collective rights.
The results showed that all participants had used social media to obtain political information. Sixty percent (60%) of them have also been actively involved in discussions and comments about the dynamics of Indonesian politics on social media. Meanwhile, 28.4 percent rarely and very rarely engage in such online discussions, and the remaining 11.6 percent chose not to answer.
Data on confirmation bias were categorized on the basis of hypothetical norms. The confirmation bias scores are grouped into three categories, low, medium, and high. Based on this grouping, it was identified that 36 participants had a high tendency of confirmation bias, while the other 43 participants were in the moderate group, and the remaining 16 participants had a low tendency of confirmation bias.
The results of the analysis show that authoritarian personality, critical openness, and perceptions of collective rights are correlated with confirmation bias. However, reflective skepticism and political preference are not correlated with confirmation bias. The results showing the insignificant relationship between political preference and confirmation bias are of interest. This actually strengthens the study conducted by Brandt, et al. (2014), who reported that liberals and conservatives tend to express the same level of intolerance towards various ideologies and threatening groups. This means that individuals can still be trapped with presumptions regardless of their political preferences.
Author: Dr. Rahkman Ardi
Details of this research can be viewed at:
http://journal.uad.ac.id/index.php/HUMANITAS/article/view/19664
Ardi, R. , & Pradiri, A.P. (2021). Determinant factors of partisans’ confirmation bias in social media. Humanitas, 18(1). https://dx.doi.org/10.26555/humanitas.v18i1.1966
