Building tools to identify low responding communities in the 2020 National Census as part of the Census Bureau's Opportunity Project.
In theory, (almost) everbody should respond to the National Census. In practice, large segments of the population do not, and this has significant costs for the jurisdictions in which those uncounted people reside.
For example, an uncounted person in California is estimated to cost ~$2000 in missed Federal funding per year. In a state with a large population, this adds up: the low response to the 1990 census is thought to have cost the state 2 billion dollars in funding over the subsequent decade (as well as a seat in the House of Representatives) [Source].
However, boosting response to the Census is not easy. We are using demographic data in order to identify different low-response areas, and the marketing strategies that they will best respond to.
We are building a mapping tool to visualize demographic variation across the USA in low-response communities, and to identify outreach strategies that will best serve those communities.
MIT.