Community Engagement in Northeast Houston, Texas: Geospatial Results from a Household Survey on the Disaster Experiences of Northeast Houston, 2021-2022

This survey was conducted as part of the "Community Engagement in Southeast Texas: Pilot Project to Enhance Community Capacity and Flood Resilience" pilot project conducted by staff at the Gulf Research Program (GRP) at the National Academies of Sciences, Engineering, and Medicine (NASEM). Building on the lessons learned through previous community engagement efforts in Southeast Texas around flood risks, this project engaged communities in Northeast Houston to explore 1) how compounding events--specifically, flooding, the Coronavirus Disease 2019 (COVID-19) pandemic, and Winter Storm Uri (2021)--increased vulnerability and risk to communities, 2) how to effectively communicate these risks to community members, and 3) how to better prepare for and mitigate these risks.
In partnership with West Street Recovery (WSR), Texas A&M University at Galveston (TAMUG), and Research 4 Progress, the Gulf Research Program (GRP) and Resilient America Program (RAP) at the National Academies of Sciences, Engineering, and Medicine (the National Academies) designed a household survey to investigate the flood-related experiences of residents from Northeast Houston using quantitative methods and probabilistic sampling. The survey, administered in December 2021-March 2022, also asked about residents' experiences with the COVID-19 pandemic and Winter Storm Uri to capture information about the compounding impacts of the pandemic and winter storm on existing flood disaster preparedness, response, and recovery.
Consultants from Research 4 Progress programmed the survey tool using Qualtrics XM, performed the survey deployment and conducted preliminary descriptive statistical analyses (e.g., descriptive statistics, cross-tabulations) of the survey data. The principal investigators then conducted an advanced statistical and geospatial analysis of the survey data. Analyses include: descriptive statistics; geocoding response using ArcGIS Pro; comparing "real" risk to perceived flood risk using a Flood Risk score created using inverse distance weighting and empirical Bayesian kriging; determining flood risk perception influence on protective action with classical and spatial regression models; and identifying risk communication preferences and types of services sought after varying types of disasters (i.e., flooding, the COVID-19 pandemic, and Winter Storm Uri) with Wilcoxon tests and contingency tables.