The American Community Survey (ACS) is a project of the U.S. Census Bureau that replaces the long form in the decennial census. It is an ongoing statistical survey, sent to approximately 250,000 addresses monthly, and thus more current than information obtained by the long form.
Many Americans found filling out the long form to be burdensome and intrusive, and its unpopularity was a factor in the declining response rate to the decennial census. In 1995, the Bureau began a process to change the means of demographic, housing, social, and economic information from the census long form to the ACS. Testing began in 1996, and the ACS program began producing test data in 2000, 2001, and 2002. The full program is expected to be implemented by 2010.
The planned sample will be 3 million housing units and group quarters in the U.S., in every county, American Indian and Alaska Native area, and Hawaiian Homeland, and in Puerto Rico annually (250,000/month). Data will be collected primarily by mail, with Census Bureau follow up.
The Department of Commerce has stated that those who receive a survey form are required to provide answers to a long list of questions about themselves and their families, including their profession, how much money they earn, their source of health insurance, their preferred mode of transportation to and from work, and the amount of money they pay for housing and utilities. Those who decline to answer these questions may receive follow-up phone calls and/or visits to their homes from Census Bureau personnel, and are threatened with prosecution and fines up to $5000. No person has ever been charged with a crime for refusing to answer the ACS survey, which several U.S. Representatives have challenged as unauthorized by the census act and violative of the Right to Financial Privacy Act. The Department of Commerce states that it is "not an enforcement agency."
The processed information will provide annual estimates for all states, as well as all cities, counties, metropolitan areas, and population groups of 65,000 people or more. For smaller areas, the Census Bureau expects useful samples to develop over time: over 3 years in areas with 20,000 to 65,000 people, and 5 years in areas with fewer than 20,000 people. The quality of these samples is expected to match that of the decennial census. However, the small area and sub-group sample is not anticipated to be benchmarked to the 2010 Census short form for cities and smaller areas, but instead to total population estimates at the county level. This will create large differences in basic count 2010 ACS estimates and the 100% 2010 United States Census counts for cities and sub-areas, and for all sub-groups (ethnic, age...) within the county.
The surveys asks for more information, and at a higher frequency of polling, than the simple enumeration required by U.S. Constitution Article I Section 2. Congressman Ron Paul of Texas, who opposes the ACS, said of it that the founding fathers of the United States "never authorized the federal government to continuously survey the American people. More importantly, they never envisioned a nation where the people would roll over and submit to every government demand."
Congressman Paul's statement regarding the ACS does, however, contain some inaccuracies regarding the distribution and sale of information from the survey. Much like the standard decennial census, detailed personal information collected by the ACS will be safeguarded for at least 75 years. While aggregate data such as race, income and age information will be freely available, much the same as the standard census; the public, including businesses, will not have access to any one person's information.