Retrospective Analysis of the Psychological Predictors of Public Health Support in Bulgarians at the Beginning of the Coronavirus Pandemic

Abstract

The earliest critical context of the pandemic, preceding the first real epidemiological wave of contagion in Bulgaria, was examined using a socio-affective perspective. A retrospective and agnostic analytical approach was adopted. Our goal was to identify traits and trends that explain public health support (PHS) of Bulgarians during the first two months of the declared state of emergency. We investigated a set of variables with a unified method within an international scientific network named the International Collaboration on Social & Moral Psychology of COVID-19 (ICSMP) in April and May 2020. A total of 733 Bulgarians participated in the study (67.3% females), with an average age of 31.8 years (SD = 11.66). Conspiracy Theories Beliefs were a significant predictor of lower PHS. Psychological Well-Being was significantly associated with Physical Contact and Anti-Corona Policy Support. Physical Contact was significantly predicted by fewer Conspiracy Theories Beliefs, higher Collective Narcissism, Open-mindedness, higher Trait Self-Control, Moral Identity, Risk Perception and Psychological Well-Being. Physical Hygiene compliance was predicted by fewer Conspiracy Theories Beliefs, Collective Narcissism, Morality-as-Cooperation, Moral Identity and Psychological Well-Being. The results revealed two polar trends of support and non-support of public health policies. The contribution of this study is in providing evidence for the affective polarization and phenomenology of (non)precarity during the outbreak of the pandemic.

Keywords: affective polarization; conspiracy theories beliefs; ontological uncertainty; precarity; psychological predictors; psychological well-being; public health support.

Figure 1

Variable importance plots from random forest models predicting Physical Contact (top row; R² = 13%), Physical Hygiene (middle row; R² = 14%) and Anti-Corona Policy Support (bottom row; R² = 20%). The ranking of variable importance is based on mean decrease in accuracy and total increase in node purity, with higher values indicating greater impact of the predictor.