Social media, quality of democracy, and citizen satisfaction with democracy in central and eastern Europe

ABSTRACT

The last decade has been tumultuous for democracy in Central and Eastern Europe (CEE). While the beginning of the 21st century saw many post-communist countries in the region democratize and become full members of the European Union, the last 10 years have been wrought with stagnation and democratic backsliding. By the 2020s, some of the strongest democratizers in the region had experienced significant issues. This study attempts to answer two questions related to whether social media users’ satisfaction with democracy relies on a country’s quality of democracy. The first question explores this relationship in the overall context of democracy, while the second examines the relationship with democratic backsliding. The findings show that social media use correlates with higher satisfaction with democracy when their country’s democracy is more robust. Furthermore, as backsliding occurs, social media use is correlated with lower satisfaction with democracy.

KEYWORDS:

• Social media
• democracy
• democratization
• public opinion

Supplementary Material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/19331681.2023.2220319.

Disclosure statement

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

Notes

1. See Table A1 in the Appendix for descriptive statistics of all dependent and independent variables.

2. The standard deviation of the original six-category variable was 2.28, whereas the variance for the three-category variable is .875.

3. For example, in 2014, Hungary’s VDEM Liberal Democracy rating is .532. In 2009, their VDEM Liberal Democracy rating was .782. By subtracting 2009’s rating from 2014 we get the Five-year change in democracy rating of -.25. Mathematically the annotation would be [2014 VDEM Liberal Democracy rating] - [2009 VDEM Liberal Democracy rating]= [Five -year change in democracy rating]. Thus, in our example .532 - .782 = -.25.

4. Full empirical results for each figure are found in Tables A3-A8 in the Appendix.

5. In all analyses, the reference category for social media use and social media news is “never use.” Thus, all other findings are in reference to people who never use SNS or never use it to seek news.

6. All empirical models in the paper are hierarchical ordinal logistic regressions because of the ordinal coding of the dependent variable.

7. As noted in the Data and Methods section, entropy balancing is performed before all the empirical models. Full weighting statistics for entropy balancing are found in Table A2 of the Appendix. The results show that entropy balancing was successful in reweighting the means of the control and treatment groups thus adding robustness to the findings in this section.

Additional information

Funding

The work was supported by the Office of the Vice President for Research, University of South Carolina .

Notes on contributors

Matthew Placek

Matthew Placek is an Associate Professor of Political Science at the University of South Carolina Upstate. His research focuses on the influence of social media and the internet on political attitudes in new democracies and non-democratic regimes. His work has been published in Democratization, East European Politics, the International Journal of Communication, and The Washington Post’s Monkey Cage Blog.