Rally Effect in the Covid-19 Pandemic: The Role of Affectedness, Fear, and Partisanship

Abstract

When the Covid-19 pandemic hit internationally in March 2020, governments and political incumbents received exceptionally high approval ratings. Such a sudden spike of public support in times of crisis is often explained as the ‘rally ‘round the flag’ effect. This paper has three goals: first, to examine whether a rally effect indeed occurred; second, to analyse whether and how much it is related to (i) affectedness, i.e. the occurrence of infections on individual and aggregate level, and (ii) fear of Covid-19; and third, to examine an assumed moderating effect of partisanship. We merged individual survey data from an online survey conducted in September 2020 as part of the German Longitudinal Election Study (GLES) with infection rates on the state level (Bundesländer) published by the Robert Koch Institute. We detect a striking rally effect in all partisan camps. Furthermore, we identify fear of Covid-19 as the driving mechanism while there is no evidence that affectedness is a major force behind the rally effect. Furthermore, we show that partisanship takes on a moderating role for fear of Covid-19 regarding satisfaction with government.

Disclosure Statement

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

Supplemental Data and Research Materials

The underlying research materials for this paper, e.g. data, samples or models, can be accessed at https://osf.io/dctr7/?view_only=4ceb8fd40c6e4b1f94e85028233962ce.

Notes

1 Mueller (1970, 21; 1973, 208) originally writes ‘involves the United States and particularly the president directly’.

2 Anger as an emotional reaction to a threatening event is associated with agency, which grows when there is an actor clearly responsible for the threat (e.g. Carver 2004). It is debatable to what extent the COVID-19 virus can be considered an abstract actor. In the beginning of the outbreak, China was blamed as the initiator of the pandemic, which was accompanied by resentment towards Chinese people (Rafi 2020).

3 The victimisation hypothesis assumes that becoming a victim of crime can happen again. Therefore, the analogy between fear of crime and Covid-19 is limited, because confirmed reinfections of the same person are presumably extremely rare. Nevertheless, persons who survive COVID-19 could still experience higher levels of fear because they worry about their friends and family.

4 In addition, we ran models for robustness checks, where we included 7-day incidences for the number of cases and deaths as long-term indicators as well as the absolute number of cases and deaths to record short-term developments during the week before the survey was conducted. Respective models yield the same results.

5 In additional analyses we included the same indicators of the occurrence of infection on a county level and ran multilevel models with three levels (individual, county, state). But again, the results for the county indicators did not display any significance.

Additional information

Notes on contributors

Melanie Dietz

Melanie Dietz is a research associate at the Research Institute Social Cohesion, Goethe University Frankfurt and a research assistant in the German Longitudinal Election Study (GLES) since 2017. In her bachelor's degree, she studied sociology and political science at the University of Mannheim, while she completed her master's degree with a focus on political sociology and social inequality at Goethe University Frankfurt. Her research interests lie in electoral behaviour, political participation, and political attitudes with a focus on gender studies.

Sigrid Roßteutscher

Sigrid Roßteutscher is a professor in the department of social sciences, Goethe University Frankfurt. She studied history and political science at the University of Mannheim and received her PhD from the European University Institute (EUI) in Florence, Italy. She acquired her post-doctoral qualification (Habilitation) at the University of Mannheim and was Co-Director of the German Longitudinal Election Study (GLES) from 2009 to 2017. Currently, she directs an international Norface programme on The Rural Urban Divide in Europe (RUDE). Her research interests focus on political participation, electoral behaviour, turnout, political inequality, the role of religion in politics, value orientations and social capital.

Philipp Scherer

Philipp Scherer is a research associate in the department of social sciences, Goethe University Frankfurt. He studied social sciences with advanced fields of study in economics and political science at the University of Koblenz-Landau from 2002 to 2008 and received his PhD from Goethe University Frankfurt in 2017. His research interests focus on electoral behaviour, political attitudes, the socio-political relevance of the left-right dimension and content analysis.

Lars-Christopher Stövsand

Lars-Christopher Stövsand is a research associate in the department of social sciences at the Goethe University Frankfurt and a research fellow in the German Longitudinal Election Study (GLES). He studied political science and history at the Christian-Albrecht University of Kiel, the University of Salamanca and Goethe University Frankfurt. His research interests focus on electoral behaviour, political attitudes, and political polarisation.