The Impact of COVID-19 on Community Engagement: Oaxaca-Blinder Decomposition between 2019 and 2021

Background/Purpose

Community engagement is central to social work practice and social welfare policy because social cohesion is closely linked to health behaviors, resilience, and emotional well-being. Prior research shows that volunteering and social cohesion reinforce each other over time, highlighting the importance of both formal and informal participation. The COVID-19 pandemic disrupted these dynamics in several ways. In-person activities declined due to public health restrictions, while perceptions of neighborhood safety and trust shifted during lockdowns. Civic and political engagement moved toward online formats, often reducing opportunities for direct interaction. Research also suggests that confidence in expressing opinions influences participation, yet individuals’ ability to contribute is often limited by unequal access to organizational resources. However, there is limited evidence on how COVID-19 affected multiple dimensions of community engagement. This study examines how engagement patterns changed from before to during/after the pandemic.

Methods

Data and samples: This study uses the California Health Interview Survey (CHIS), a large, population-representative survey of over 20,000 households per cycle. CHIS employs geographically stratified, address-based sampling to represent California’s population. This study combines the 2019 adult data from 2019-2020 cycle (“pre-COVID”) and 2021 adult data from 2021-2022 cycle (“during/after-COVID”) to assess the structural impact of COVID-19 on poverty and employment pattern between 2019 and 2021.

Measures: COVID-19 exposure is proxied by survey timing. Social cohesion is measured using a shortened Social Cohesion and Trust Scale (Sampson et al., 1997). Additional measures include volunteering, community political group membership, perceived neighborhood safety, confidence expressing views, and ability to contact elected officials, mostly coded dichotomously.

Analysis: This study uses Oaxaca-Blinder decomposition (Blinder, 1973; Oaxaca, 1973; Jann, 2008) to assess the overall impact of COVID-19 on poverty and employment. Oaxaca-Blinder decomposition is used to assess outcome differences between 2019 and 2021. This approach decomposes outcome differences into 1) portions explained by changes in observed characteristics (“endowments”) and 2) portions due to changes in how these characteristics are related with outcomes (“coefficients”), isolating COVID-related structural effects.

Results

Compared with 2019, respondents in 2021 reported a small but statistically significant increase in social cohesion (8.93 vs. 8.80, p<0.001), suggesting stronger perceived solidarity despite the pandemic. In contrast, formal participation declined: volunteering decreased (10.9% vs. 12.7%, p<0.001), and the likelihood of feeling completely safe in one’s neighborhood also declined (33.8% vs. 35.5%, p=0.015). Differences in community political group membership, broader safety measures, and confidence expressing views were not statistically significant. Willingness to contact elected officials was slightly higher in 2021 but only marginally significant. For these significant outcomes, the gaps were driven mainly by coefficient effects rather than compositional differences, suggesting that COVID-19 has affected community engagement pattern.

Conclusion and Implications

The findings suggest that community life in 2021 was reshaped rather than uniformly weakened by COVID-19. Notably, social cohesion increased despite the pandemic, even as formal volunteering and the strongest sense of neighborhood safety declined. This points to the importance of policies and interventions that sustain informal solidarity while rebuilding opportunities for organized participation, civic action, and neighborhood trust after COVID-19 era.