ASFPM Policy Priorities

Big Data

What It Is

Large national datasets encompass a wide range of data collected at the national level by government agencies, research institutions, and other organizations. These datasets typically cover various aspects of a nation’s demographics, economy, environment, infrastructure, public health, and more.  These datasets are the foundation of indices and tools and can be invaluable resources for better risk awareness, preparing and planning for future events, disaster mitigation efforts, and even prioritizing and allocating resources, but they also come with their own set of challenges.

Why It Matters

• Large datasets often comprise data from multiple sources, each with its own format, quality standards, and levels of completeness. Oftentimes these datasets lack sufficient granularity to truly understand the risk to communities and individuals. Integrating these heterogeneous datasets can be challenging and may lead to inconsistencies, inaccuracies, or significant data gaps, leading to concerns about data quality and completeness.  
• National datasets typically originate from various government agencies, research institutions, and private organizations, each collecting data for specific purposes. Integrating these disparate datasets into a cohesive framework can be complex due to differences in data formats, schemas, data density and access protocols. Ensuring interoperability among different systems and databases is crucial to harness the full potential of these datasets.
• National datasets may reflect underlying biases in data collection methods, sampling strategies, or demographic distributions. Biases in the data can lead to disparities in disaster risk assessments, response strategies, and resource allocation, often disproportionately affecting marginalized communities or regions. Addressing these biases requires open vetting of the methodology used to create these datasets, careful consideration of data sources, validation techniques, consideration of other more granular data sources when available, and equitable representation to ensure fair and effective disaster mitigation policies and interventions.

Our Position

There needs to be a clear-eyed view of the promises and perils of big data. Data density varies across the country. While large scale datasets can be appropriate for assessment and planning, they may not be appropriate to make equitable funding priorities, determination of cost-share or be relied on for construction/design purposes. National datasets must allow for supplementation from better local or state data, where available, to ensure a sufficient level of data density and integrity, and equitable distribution of available funding.

Key Resources on Big Data

• One
• Two
• Three