Big Data (2024)

Big Data's mission is two-fold: 1) to apply cutting-edge data collection and analytics techniques to ensure the U.S. Census Bureau can maintain its high quality data products in a timely and efficient manner in the 21st Century; and 2) to create new products for public and business users. Our focus is on using data from both outside and inside the U.S. Census Bureau to improve current statistical products, provide new opportunities for research and innovation, and create new products and processes while reducing costs.

Big data is a term used to describe data sources that are fast-changing, large in both size and breadth of information, and come from sources other than surveys. Examples include retail and payroll transactions, satellite images, and "smart" devices. Big data also includes administrative data from federal, state, and local governments, as well as third party providers. Typically, big data is “found” or “observed”, in that it is collected passively as the digital exhausts of personal and commercial activities. Such seemingly disparate data sources and techniques can provide unique insights that were not easily observable previously. Combining big data from different sources for the U.S. Census Bureau’s research purposes and product development requires utilization of new techniques, including machine learning.

The U.S. Census Bureau is charged with gathering accurate information on the U.S. population and economy. By utilizing big data sources and techniques, we can use our existing data more efficiently, and generate new insights previously beyond the capabilities of both the government and private sector. Using the U.S. Census Bureau's unique position in the Federal Statistical System, big data allows us to combine and create data products from government agencies and private sources for the first time, creating a new wave of statistical products and economic insights. The use of big data sources and techniques allows us to generate more accurate, granular and timely statistics at lower costs, helping the U.S. Census Bureau transform itself into a statistical agency for the 21st Century.

Examples:

Working with partners inside and outside of the federal government, we are applying big data research techniques to:

1. Study the impact of the “gig” economy.
2. Improve and update the NAICS classification of all U.S. businesses, on a yearly basis, at a lower cost.
3. Decrease the cost of survey operations, using predictive models to train and help field representatives.
4. Identify and improve healthcare outcomes.
5. Demonstrate how research funding at universities impacts the local economy and career outcomes of students.