Keep that mask on: will Germans become more like East Asians?

The study investigates whether increased mask usage observed during COVID-19 will persist in Germany after the pandemic. Early 2020 messaging on mask efficacy was mixed globally, but accumulating evidence and policy changes (e.g., Germany’s nationwide mask requirement by late April 2020) shifted behaviors and norms. Pre-pandemic, mask-wearing in Germany was rare compared to East Asian countries where masks were commonly used to prevent respiratory disease transmission. The research question asks if Germans plan to continue wearing masks post-pandemic and whether attitudes about learning from East Asia predict such intentions. The study’s importance lies in potential public health and economic benefits if protective routines persist beyond COVID-19.

Initial experimental and observational evidence during 2020 supported mask effectiveness (e.g., Chen et al., 2020; Yang et al., 2020; Zhang et al., 2020; Mitze et al., 2020), and a meta-analysis by Chu et al. (2020) highlighted masks’ role in preventing spread. In Germany, acceptance evolved from skepticism—many found masks ‘strange’ (Rieger, 2020)—to a new social norm, with willingness among social multipliers to sanction non-compliance (Schunk and Wagner, 2020). East Asians often adopted masks based on prior epidemic experience, even outside East Asia (Sit et al., 2020). Policy responses documented by OxCGRT (Hale et al., 2020) show varying stringency across countries. Prior work also documented planned behavioral changes during COVID-19 (Matschke and Rieger, 2021). This context motivates examining whether Germans might adopt mask-wearing habits more akin to East Asian practices after COVID-19, and whether the belief that one can ‘learn from East Asia’ correlates with mask attitudes and intentions.

Large-scale online survey conducted by Trier University as part of a broader panel (>3000 participants) starting March 2020 (Rieger and He-Ulbricht, 2020). The December 2020 wave (N=656) first asked about planned mask-wearing after the pandemic across five scenarios: (1) fully occupied bus during flu season; (2) crowded indoor event during flu season; (3) fully occupied bus with own cold symptoms; (4) crowded indoor event with own cold symptoms; (5) visiting an older, frail person with own cold symptoms. Responses on a 4-point Likert scale (1=totally disagree to 4=totally agree). Additional items from previous waves included: perceived strangeness of mask-wearing and agreement that ‘we can learn a lot from East Asia when it comes to dealing with epidemics’ (same 4-point scale); perceived protection by masks for self and for others (5-point scale: 1=not at all to 5=very well); and an index of mask-wearing frequency during the pandemic. A follow-up in January 2021 re-asked the five planned-behavior items in a subsample (N=134) to test stability. Sample characteristics reflect Trier University’s focus on humanities/social sciences: mainly students, majority female; thus not representative of the German population. Constructed a ‘willingness to wear masks after COVID-19’ factor combining the five scenarios (Cronbach’s alpha=0.91). Statistical analyses included descriptive statistics, Pearson correlations, and OLS regressions with standardized variables. Regressions examined determinants of post-pandemic willingness (demographics: age, gender, student status, bachelor’s degree; perceived protection for self and others; belief in learning from East Asia). Subsample analyses split by gender and age (<25 vs. ≥25 years). Model fit reported via adjusted R² and significance via t-values.

Descriptive intentions (December 2020, N=656): - Fully occupied bus during flu season: 52.7% agree/fully agree (Mean=2.54, SD=0.95; 15.5% totally disagree, 31.7% rather disagree, 35.8% rather agree, 16.8% totally agree). - Crowded indoor event during flu season: ~32.6% agree/fully agree (Mean=2.15, SD=0.90; 25.8% totally disagree, 41.6% rather disagree, 24.2% rather agree, 8.4% totally agree). - Fully occupied bus with own cold symptoms: ~69.4% agree/fully agree (Mean=2.90, SD=0.98; 10.1% totally disagree, 20.9% rather disagree, 35.7% rather agree, 33.7% totally agree). - Crowded indoor event with own cold symptoms: ~60.1% agree/fully agree (Mean=2.73, SD=0.89; 13.9% totally disagree, 26.1% rather disagree, 33.7% rather agree, 26.4% totally agree). - Visiting an older, frail person with own cold symptoms: 86.2% agree/fully agree (Mean=3.38, SD=0.89; 7.2% totally disagree, 6.6% rather disagree, 27.4% rather agree, 58.8% totally agree). Stability over time: In January 2021 re-survey (N=134), mean responses were similar to December 2020 and not statistically different overall; for visiting an older, frail person while symptomatic, willingness increased (mean 3.61 vs. 3.38 in December). Regression analyses (OLS, standardized variables): - Factor ‘willingness to wear masks after COVID-19’ shows high internal consistency (Cronbach’s alpha=0.91). - In models splitting by gender and age (Table 2), demographic variables (age, gender, student status) were not statistically significant with small coefficients; perceived protection of others by masks was a strong, highly significant positive determinant of willingness. Adjusted R² ranged ~7.2%–15.3%; sample sizes by specification N≈275–366. Correlation analyses (Table 3; Pearson correlations): - ‘Learning from East Asia’ correlates positively with willingness to wear masks after COVID-19 (r=0.201, p<0.01, N=656), current mask-wearing (r=0.231, p<0.01, N=1162), perceived self-protection (r=0.315, p<0.01, N=1035), and perceived protection of others (r=0.192, p<0.01, N=1355); it correlates negatively with viewing masks as strange (r=−0.101, p<0.01, N=500). Learning from East Asia as predictor (Table 4): - ‘Learning from East Asia’ is a significant predictor of post-pandemic mask-wearing intentions across models (p<0.001 where included), even controlling for perceived protection and demographics. The effect size is smaller than that of perceived protection of others but remains robust. Adjusted R² improves when including the East Asia variable (e.g., up to ~13.6%); N up to 643 depending on available variables. Perceived protection determinants (Table 5): - ‘Learning from East Asia’ significantly predicts higher perceived protection by masks (standardized coefficient ≈0.135, p<0.05); demographics not significant; Adj. R²≈1.8% (N=1322). Interpretation: Intentions are stronger in altruistic contexts (protecting others), e.g., with own symptoms or visiting vulnerable individuals, suggesting internalization of public health messaging that masks primarily protect others.

Findings indicate that many Germans plan to continue wearing masks in specific contexts after the pandemic, especially when symptomatic or around vulnerable individuals. This marks a shift from pre-COVID-19 norms and aligns more with practices common in East Asia. The strongest driver of intentions is the belief that masks protect others, highlighting altruistic motivation and the effectiveness of public health messaging. Additionally, the perception that Germany can learn from East Asia correlates with and predicts pro-mask attitudes and planned behaviors, even when controlling for perceived efficacy and demographics. These results suggest evolving social norms toward situational mask use and imply potential public health gains if such behaviors persist, particularly during respiratory virus seasons or when symptomatic.

Survey evidence suggests a sustained, situational adoption of mask-wearing in Germany beyond COVID-19, especially in high-risk settings (crowded transport or events during flu season) and when symptomatic or visiting vulnerable individuals. Such behavioral carryover could reduce morbidity and mortality from common respiratory illnesses and yield economic benefits by lowering healthcare utilization and productivity losses associated with colds and influenza. More broadly, attitudes indicating that Germans can learn from East Asia are linked to intentions to continue mask use, implying that cross-cultural learning may facilitate durable protective routines. Future research could use representative samples, longitudinal behavioral tracking, and experimental designs to assess causal pathways and real-world adherence post-mandate.

The sample is not representative of the German population (primarily students; majority female; single university context), limiting generalizability. Intentions may not translate into future behaviors, introducing potential intention–behavior gaps. Not all variables were collected in every survey wave, leading to varying sample sizes across analyses and limiting inclusion of all covariates simultaneously. Self-reported measures of perceived protection and behavior may be subject to social desirability and recall biases.