Social cohesion—defined as the strength of relationships and sense of solidarity among members of a community—is increasingly recognized as a core component of disaster resilience. It contributes to how communities prepare for, respond to, and recover from disruptive events by fostering trust, information sharing, mutual aid, and collective action.Reference Portes and Vickstrom1, Reference Schiefer and Van der Noll2 In disaster contexts, these elements of cohesion often determine the effectiveness of informal response networks and the speed of community recovery. As Schoch-Spana et al. (2019) note, resilience is not merely a function of infrastructure or emergency planning but is deeply rooted in the social dynamics that enable communities to function under stress.Reference Schoch-Spana, Gill and Hosangadi3
Research suggests that socially cohesive communities are better equipped to respond to crisis. For example, cohesion has been linked to increased volunteerism, higher levels of trust in leadership, and more rapid mobilization of local resources.Reference Schoch-Spana, Gill and Hosangadi3, Reference Townshend, Awosoga and Kulig4 Despite this growing recognition, the mechanisms by which cohesion influences recovery, and how this relationship varies by context, remain areas of investigation.
This study explores directly the association between social cohesion and community resilience in the aftermath of the 2015 Memorial Day Flood in Hays County, TX. The event, which caused catastrophic flooding along the Blanco and San Marcos rivers, resulted in 13 fatalities and destroyed hundreds of homes. In its wake, local volunteer efforts, neighbor-to-neighbor assistance, and grassroots recovery efforts provided anecdotal evidence of cohesion-driven resilience.
Methods
An adapted Community Assessment for Public Health Emergency Response (CASPER) methodology to conduct a household-based survey in Hays County, TX was used. CASPER is a 2-stage cluster sampling technique developed by the Centers for Disease Control and Prevention to rapidly assess community needs after a disaster. This method has been successfully used in previous hazard events, including an industrial fire in the Houston Ship Channel.Reference Sansom, Aarvig and Sansom5 Thirty clusters were randomly selected across affected areas in San Marcos and Wimberley, with households systematically sampled within each cluster. Data were collected during the summer and fall of 2022 by teams from Texas A&M University’s EpiAssist program, working alongside community leaders from Blanco River Academy. Each team included at least 1 bilingual member to support Spanish-speaking households when needed.
The survey included demographic and disaster impact questions—such as receipt of hazard warnings, household damage, length of community residency, and perceived recovery progress. To assess social cohesion, we used Buckner’s Index of Cohesion (BIC), a validated 18-item scale measuring psychological sense of community, degree of neighboring, and attraction to neighborhood.Reference Townshend, Awosoga and Kulig4, Reference Buckner6 Community resilience was measured using the 11-item Index of Perceived Community Resilience (IPCR), which includes subscales on leadership and empowerment, community engagement, and adverse geography.Reference Kulig, Edge and Townshend7 These tools have previously been applied in disaster contexts to explore the relationship between social connectivity and adaptive capacity.Reference Townshend, Awosoga and Kulig4, Reference Kulig, Edge and Townshend7
Results
A total of 128 participants completed the survey (n=128), yielding a response rate of 72.2%. Females slightly outnumbered males, comprising 53.91% and 46.09% of respondents, respectively. The average age was 52.76 years (SD = 16.48). Most participants identified as White (77.95%), with additional representation from Black (5.51%), Other (13.39%), Asian, and American Indian respondents. Approximately 26.56% identified as Hispanic. Education levels varied, though 25% held a bachelor’s degree, and over 43% had at least some college or an associate’s degree. Household incomes were distributed across 4 main brackets, with the largest share falling between $32 048 and $106 827.
The scatterplot analysis of Buckner’s Index of Cohesion (BIC) and the Index of Perceived Community Resilience (IPCR) revealed a modest but statistically significant relationship (R² = 0.2007, P < 0.001), as shown in Figure 1. This suggests that higher community cohesion is associated with higher perceived resilience, though other contributing factors remain. Multiple linear regression (Table 1) further confirmed the significant positive association between BIC and IPCR (coef = 0.25, P < 0.001). Other variables—including sex, race, age, income, tenure, and political ideology—did not demonstrate significant associations with IPCR, highlighting the central role of community cohesion in shaping perceived resilience outcomes.
Index of Perceived Community Resilience (IPCR) Composite score by Buckner’s Index of Cohesion (BIC).
Multiple linear regression comparing the covariates BIC, sex, race, age, tenure in neighborhood, income, and political ideology on IPCR
Discussion
Our study found a significant positive association between community cohesion and perceived resilience, while demographic variables (e.g., age, gender, socioeconomic status) showed no significant relationships. These findings highlight the role of strong social ties in disaster recovery, supporting previous research that links social networks to resilienceReference Magsino8 and echoing recent work demonstrating similar outcomes in cohesive communities.Reference Hawken, Sunindijo and Sanderson9 However, limitations should be noted: the sample size was relatively small, predominantly non-Hispanic white, the survey occurred years after the disaster (potentially affecting recall), and all responses were self-reported, which may affect generalizability. Future research should focus on longitudinal data collection.
From a practical perspective, these findings suggest several policy-level applications for public health and emergency preparedness practitioners. Programs and plans aiming to build disaster-resilient communities should explicitly include strategies to strengthen social cohesion. Practitioners should consider several key actions. First, they can integrate cohesion metrics—such as community trust, volunteerism rates, and civic engagement levels—into community risk assessments and resilience planning frameworks. Second, they should invest in programs that promote neighborhood connectivity, including mutual aid groups, regular community meetings, and public events. These initiatives help build the “social infrastructure” critical during times of crisis.Reference Aldrich and Meyer10 Finally, practitioners should engage grassroots organizations, faith-based groups, and other trusted local networks in emergency preparedness and response planning to improve outreach, communication, and the equitable distribution of resources when disasters occur. By embedding social cohesion into emergency preparedness, practitioners can strengthen community resilience more holistically. A city with strong social ties and mutual trust lays the groundwork for resilience, but true resilience also relies on structural and resource-based factors. Our results suggest that cohesion enhances disaster resilience, though it does not guarantee it.
Data availability statement
Data are available upon reasonable request.
Author contribution
-
• Garett T. Sansom: Conceptualization, methodology, writing—original draft, supervision.
-
• Lindsay Sansom: Community engagement coordination, writing, review, and editing.
-
• Lyssa Losa: Data collection and survey implementation.
-
• Ruby Hernandez: Data collection and survey implementation.
-
• Jacquita N. Johnson: Data collection and survey implementation.
-
• Benika Dixon: Statistical review and writing—methods and discussion.
Funding statement
Funding for this study came from the Texas A&M University Superfund Research Center (National Institute of Environmental Health Sciences P42ES027704-01 Center) and the National Science Foundation (NSF) under Grant # 1931301.
Competing interests
The authors declare no conflicts of interest.
Ethical standard
This project and all related materials were approved by the Texas A&M University Institutional Review Board (IRB2021-1078).
