Feeding America datasets - onetomapanalytics/Meta_Data GitHub Wiki
Feeding America datasets
General description
- Database primary purpose - To improve the understanding of how food insecurity and food costs vary at the local level aiming to provide knowledge so communities can develop more targeted strategies to reach people struggling with hunger
- Overall data type - Demographics
- Dataset type - longitudinal
- Data source - Current Population Survey (CPS), American Community Survey (ACS), and Bureau of Labor Statistics (BLS)
- Data level - county- and congressional district-level
- Geographic location of the data collection sites - United States
- Sponsor, manager, or home institution - Feeding America
- Date range - 2011-2022: Map the Meal Gap; 2019-2020: State of Senior Hunger
- Geolocation data - FIPS code, state, county, district
- Dates - Year
- Longitudinal tracking - Track state, county, or district
- Financial variables - Cost per meal
- Variables that are uniquely present in this dataset - Local food insecurity estimates among all individuals and children by income category and local food expenditure estimates among people who are food insecure and food secure
Applicable methods
- Association methods, such as logistic regression (1, 2), linear regressions (3), random effects models (4)
- Bivariate analysis (5)
- Machine learning (6)
- Spatial linear modeling (7)
High-impact designs
- To determine if greater non-profit hospital spending for community benefits is associated with better health outcomes in the county where they are located (8)
Data dictionary
To access the Feeding America datasets data dictionary, click here
Variable categories
- Geolocation variables (i.e., state, county, district, FIPS code)
- Demographics (e.g., number of food insecure persons by ethnicities, number of food insecure children)
- Food insecure rates (e.g., overall, for children, and for older adults)
- Financial (e.g., cost per meal, weighted weekly money needed by a food-insecure person)
Linkage to other datasets
- Linkages can be established for any dataset that might have geolocation data (i.e., FIPS code)