Impact
To the extent that the census undercounts minority populations, it dilutes minority voting rights and reduces the federal funds available to those communities in which the undercounted minorities live and perpetuates the chronic shortage of government services available to populations that are disproportionately low income. Ensuring a fair and accurate census is thus one of our most significant civil rights challenges.
Census data directly affects decisions made on a great number of matters of national and local importance, including education, employment, veterans' services, public health care, rural development, redistricting, the environment, transportation, and housing.
- Many federal programs use census data to conduct their daily operations.
- The federal government uses census figures to allocate more than $180 billion in grants to states for, among other things, education, construction, health programs, and crime programs.
- States also use census figures to allot money within their borders, including monies for direct government services.
- Congressional seats are reapportioned and legislative districts are drawn based on Census data.
- Census data is used to monitor compliance with civil rights statutes and anti- discrimination laws.
While the overall accuracy of the census steadily improved for decades, in 1990, however, this trend was reversed and the overall accuracy of the 1990 Census actually declined from earlier years.
In 1990, approximately:
- 8.4 million people were missed all together
- 4.4 million were counted twice
- 13 million were put in the wrong place/category
Background
Following the 1990 Census, the Census Bureau, professional statisticians, and Congress agreed that significant changes were required for the 2000 census. In 1991, bipartisan legislation passed unanimously by Congress and signed into laws by President Bush directed the Secretary of Commerce and the National Academy of Sciences to analyze the census and figure out how to obtain the greatest accuracy.
Five years later, in 1996, The Plan for Census 2000, was issued.
- More basic, readable form unveiled
- Better accessibility in public places
- More effective advertising
- More questionnaires were sent to those households who didn't previously respond
- New sampling techniques were implemented to try to end the undercounting of the poor, people of color, and children.
While the plan found statistical sampling to be a valid means of reducing the undercount to an acceptable level. Opponents of sampling argued that the Constitutional language calling for "actual enumeration" implied the strict use of physically counting people.
Changes made in 1997:
- The Senate bill included $354 million for Census 2000 activities
- The Census Bureau was prohibited from making "irreversible" plans to use sampling
- Compromise reached the last day of the 1997 Congressional session:
- Creation of a Census Monitoring Board
- The Census Bureau was required to produce 2 sets of results, one using the sampling, the other not.
- A 1998 Census Dress Rehearsal tested both methods, with the rehearsals in Sacramento, CA and the Menominee Indian Reservation in Wisconsin using scientific sampling following a direct enumeration effort, while the Columbia, SC used direct enumeration only.
January 1999:
- U.S. Dept. of Commerce vs. U.S. House of Representatives: Supreme Court held the 1976 Census Act prohibited use of sampling to count population for purpose of apportioning Congressional seats.
- The Census Bureau requested $1.7 billion more in funding and had to revise plans for the 2000 Census the following year
