American Community Survey FAQ
Frequently Asked Questions

Purpose | Scope | Content | Operations | Results | American Community Survey

Purpose

What is the American Community Survey?
Short answer:
The ACS is an ongoing survey that provides an updated portrait of America every year, replacing the Census Bureau's "long form" that produced data only once every 10 years It's a key part of the Census Bureau's efforts to meet the American public's demand for more timely census data.
Longer answer:
The country needs fresh census data more than once every ten years. The census "long form" was like a snapshot of community characteristics taken once every ten years. But during the ten years between censuses that snapshot fades, and the data becomes outdated and irrelevant. The American Community Survey's new, innovative statistical methods allow the Census Bureau to produce data every year. The annual data is like a moving picture that gives shows trends and changes in community characteristics from one year to the next, giving community planners the tools they need to make informed decisions.

Top

Scope

How many addresses get the survey?
Beginning in January 2005, the ACS

How does the sample size compare to Census 2000?
Approximately 1-in-6 addresses received the "long form" in Census 2000, and about 1-in-8 addresses will receive the ACS over the next five years, which is the length of time it will take to accumulate statistically accurate data for even the smallest geographical areas.

How often will I get the survey?
No address will receive the survey more than once every five years, and some might not get it for decades, if ever.

How many addresses are included in the survey?
About 3 million addresses per year beginning in 2005, or 250,000 per month.

How many counties are in the U.S.?
There are 3,141 counties and county equivalents in the 50 States and the District of Columbia. They are categorized as follows:

Do you have data for metropolitan statistical areas (MSA's)?
Yes - currently for most MSA's of 250,000 or more

When will you begin surveying Puerto Rico?
The "Puerto Rico Community Survey" began in January 2005

When will you begin surveying group quarters and institutions?
Group quarters data collection begins in 2006, with data included in the August '07 release.

What are "group quarters and institutions?"
Group Quarters Population -- The group quarters population includes all people not living in households. Two general categories of people in group quarters are recognized: 1) the institutionalized population which includes people under formally authorized, supervised care or custody in institutions at the time of enumeration (such as correctional institutions, nursing homes, and juvenile institutions) and 2) the noninstitutionalized population which includes all people who live in group quarters other than institutions (such as college dormitories, military quarters, and group homes).

What is the timeline for getting results for smaller areas?
See chart:
Chart showing the  timeline for the release of data.
* Census tracts are small, relatively permanent statistical subdivisions of a county averaging about 4,000 inhabitants. Census block groups generally contain between 600 and 3,000 people. The smallest geographic level for which data will be produced is the block group; the Census Bureau will not publish estimates for small numbers of people if there is a probability that an individual can be identified.

What percentage of the population, will be covered cumulatively under the 65,000 and 20,000 thresholds?
In terms of counties, 82.5% of the US population live in counties of 65,000 or greater, 95.5% live in counties of 20,000 or greater

Top

Content

Number of questions?
There are more than 60 questions on the survey, and they are essentially the same as the questions that were on the Census 2000 long form

What is the wording of the survey questions?
Go to the online media toolkit - accessible via the ACS web page (www.census.gov/acs) to see a PDF copy of the survey

What is the subject definition of each of the questions?
The Census Bureau Web site provides a list of subject definitions.

Why do you ask these questions?
The Census Bureau is the leading source of quality data about the nation's people and economy. All questions are required by law for various programs. Background on each question is provided at: http://www.census.gov/acs/www/SBasics/SQuest/fact.htm

How long does it take to complete the survey?
Approximately 38 minutes.

Will there still be a ten-year census?
Yes - but the ten-year census will now focus solely on the "short form" which is the official, constitutionally mandated population count.

Top

Operations

How do I know this is a legitimate survey, or that the person at my door is a bona fide census worker?
Any request for survey information from the Census Bureau will be clearly identified as coming from the U.S. Census Bureau and as OFFICIAL BUSINESS of the United States. If a Census Bureau employee is at your door, they will: If you have questions about the survey, you may call or e-mail the Census Bureau Regional Office conducting the survey in your area. That number can be obtained from the Census Bureau Web site or by calling 301-763-INFO (4636).

When did the ACS start?
The ACS has been tested in select pilot sites around the country since 1996. Since 2000 it has been in every county and city with populations of 250,000 or more. It has been in every county nationwide since January 2005.

How do you conduct the survey?
The survey is conducted using a mailed form. Non-respondents may be contacted by telephone and/or a personal visit. http://www.census.gov/acs/www/SBasics/DataColl.htm

What is the ACS response rate?
Overall US response rate in 2003 was 96.7 percent. Rates for states can be found at http://www.census.gov/acs/www/UseData/sse/index.htm. This is the response that includes not only mail response, but phone and in-person interview follow-up. (Go to ACS/Using the Data)

How do you select the addresses?
The addresses are randomly selected to ensure a statistically valid sample.

What is the cost of non-response follow-up?
It costs about 10 times as much to obtain a response via a personal visit than it does via initial mail response.

How long between receiving the survey in the mail and getting a follow-up call or visit?
About 6 weeks.

Is response mandatory?
Yes, just as you are required to respond to jury duty, you are also required to respond to the survey should you be randomly selected to do so.

From Legal: Pursuant to Title 13 U.S.C., Sections 141, 193, and 221, the Census Bureau is authorized to both conduct the decennial census and collect related preliminary and supplementary information. The ACS form, now used each month throughout the decade, is the substitute for the census long form which has been used in previous decennial censuses. The census long form, an authorized part of the every ten year census, has been sent to approximately one out of every six households, and is virtually identical to the ACS form in numbers and types of questions. Thus, since the ACS is the substitute for the legally required census long form, the Census Bureau with Office of Management and Budget , and Congressional approval, uses the mandatory provisions of Title 13 U.S.C. to collect the monthly ACS data.

What is the penalty for nonresponse?
The Census Bureau is not a prosecuting agency; our goal is to gather and return to the public quality information about communities and those who provide goods and services. If a fine were to be applied, it would be at the discretion of the Department of Justice and would be $100 to a maximum of $5,000.

How does the cost of ACS compare to the cost of the decennial long form?
ACS is cost-neutral. Even though we are surveying on an ongoing basis rather than once every 10 years, we have spread the workload and the burden of the 10-year long form over the course of ten years.

How much is ACS costing?
The appropriation in 2004, when the sample size was 800,000 addresses, was $64 million. In 2005, with the survey at an annual sample of 3 million addresses, the appropriation is $146 million. The President's FY 2006 budget requested $169.9 million for ACS, which will allow the collection of Group Quarters data.

Top

Results

When are ACS results released?
Results are released during the summer after the survey's calendar year.

What does a 90% confidence interval mean / lower & upper bounds mean?
A 90% confidence interval is a statistical term that means there is 90% certainty that the true answer lies within the listed upper and lower percentages. It is based on mathematical models used for statistics. For more information see A Basic Explanation of Confidence Intervals http://www.census.gov/hhes/www/saipe/techdoc/stcty/ci.html

What are the data used for?
Why should I respond?
What's in it for me?
Answering the American Community Survey directly affects the amount of Federal and state funding communities receive for neighborhood improvements, public health, education, transportation and much more. Roughly $200 billion is allocated to state and local areas based on long-form data that is currently updated only every ten years.