Social media use and polarized redistributive attitudes: a comparative and causal perspective

ABSTRACT

As social media has become a means of information sharing, knowledge dissemination, and social interactions, it follows that its use significantly influences the formation of attitudes and social norms. Existing literature has raised significant concerns about social media causing political polarization. This study examines whether the use of social media causes polarized attitudes on redistribution by applying the instrumental variable (IV) estimation to the Asian Barometer Survey (14 countries). We find that informational and expressive social media use contributes to citizens’ diverging redistributive attitudes, and these effects are more pronounced in countries that are perceived as less democratic. Based on our findings, we discuss policy implications regarding the role of social media in forming redistributive attitudes.

KEYWORDS:

• Social media
• political polarization
• redistributive attitudes
• econometric approach
• instrumental variable estimation
• comparative politics

Disclosure statement

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

Notes

1 There are two distinct forms of political polarization: ideological and affective (Iyengar et al., 2019). This study directly investigates ideological polarization by focusing on divergences in social attitudes but also links to affective polarization when discussing the possibility of social conflicts and antagonism following changes in social attitudes. Thus, we group both concepts under the umbrella of political polarization.

2 Note democracy levels are measured based on citizens’ perceptions and the subjective evaluation of democracy. The evaluation of democracy can be examined using the individual-level results. However, in this case, attitudes toward democracy may be closely related to redistributive attitudes, resulting in an endogeneity problem. Considering that the main goal of the study is to examine the level of democracy in a country from a comparative political point of view, this study uses the national average value of each country.

3 Attrition bias may appear by excluding samples that do not respond to every question. However, the results did not change significantly when we included these samples.

4 When we denote the explanatory variable as X and instrumental variable as Z, the conditions imply that corr(X, Z) ≠ 0 and corr(Z, u) = 0, respectively.

5 Still, estimation results may be confounded. For example, internet access may be conditional on individual wealth levels; thus, users may have more extreme attitudes about redistribution (although this possibility is low considering that internet access is established at the national level and the term of country fixed effects in the regression can capture this effect). Thus, we should be cautious when interpreting the results as evidence of clear causality.

6 Appendix 1 shows the OLS estimation results for Hypothesis 1. Appendices 2 and 3 provide the first-stage estimation results for Hypotheses 1 and 2.

7 With an exception, redistributive attitudes measured by an equal treatment deviation are polarized when household income is high, consistent with our understanding that a wealth-transfer policy affects the rich and poor differently.

Additional information

Notes on contributors

Hoyong Jung

Hoyong Jung is an associate professor in the department of economics, Kookmin University, South Korea. He holds Ph.D. in economics at the University of Wisconsin-Madison. His research has been published at Party Politics, International Political Science Review, Political Studies Review, and among others.

Sangwon Lee

Sangwon Lee is an assistant professor in the Department of Communication Studies and an Affiliate Faculty member of the Data Science and Applications Center at New Mexico State University. His research examines how social media use changes the ways in which people learn about, engage with, and form opinions on political and social issues. His research has appeared in Communication Research, New Media & Society, Human Communication Research, Mass Communication and Society, and Computers in Human Behavior among others.