American Community Survey Overview Jerry Wong Information Services Specialist Los Angeles Regional Office U.S. Census Bureau 1/10
What is the American Community Survey? A large, continuous survey that: • is sent to 3 million resident addresses per year • produces characteristics of population and housing • produces estimates for small areas and small population groups • Produces characteristics, not a population count • Key component of the decennial census program
History and Timeline of the ACS Congress renewed interest in an alternative to the once-a-decade census First operational test conducted in 4 sites N=866,000 housing units in 1,239 counties First 1-yr estimates released for areas 65,000+ Continuous measurement conceptualized N=165,000 housing units 2009 1981 1990 1994 2006 2000 2005 1995 1999 2008 Work began on what evolved into the ACS Test sites expanded yearly to 36 counties in 26 states Group Quarters added N=3,000,000 housing units in all counties in 50 states, DC & PR First 3-yr estimates released for areas 20,000+
Developing the ACS The ACS was developed to: • Focus the Census on improving the population count • Provide characteristic data more than once every 10 years to frame policy issues • Allow use of current data to respond to new trends
Decennial Census Census 2000 used 2 forms • “short” form – asked for basic demographic and housing information, such as age, sex, race, how many people lived in the housing unit, and if the housing unit was owned or rented by the resident • “long” form – collected the same information as the short form but also collected more in-depth information such as income, education, and language spoken at home Only a small portion of the population, called a sample, received the long form.
Census 2000 and ACSSimilarities • Many questions similar • Many of the same basic statistics are released • Comparisons can be made for most population and housing subjects • http://www.census.gov/acs/www/UseData/compACS.html • 5-year estimates will be produced for same broad set of geographic areas • including census tracts and block groups, zip code tabulation areas
Census 2000 and ACSKey Differences ACS data now available for larger and mid-sized areas ACS data for small geographic areas and population groups will be produced every year starting in 2010 ACS data describe a period of time and published data are based on 12 months, 36 months, or 60 months
Census 2000 and ACSData Quality • Goal of the ACS: Produce data of comparable quality to Census 2000 long form • ACS 5-year data • Sampling error is larger in ACS due to smaller sample size • Non-sampling error is smaller due to: • Lower rate of nonresponse • Higher item response rates • Permanent interviewing staff using automated data collection
Social Characteristics Education Marital Status Fertility Grandparent Caregivers Veterans Disability Status • Place of Birth • Citizenship • Year of Entry • Language Spoken at Home • Ancestry/Tribal • Affiliation 11 11
Economic Characteristics Income Benefits Employment Status Occupation Industry Commuting to Work Place of Work 12 12
Housing Characteristics Tenure (own vs. rent) Occupancy & Structure Housing Value Taxes & Insurance Utilities Mortgage/Monthly Rent 13 13
Demographic Characteristics Sex Age Race Hispanic Origin 14 14
Recent Content Changes • New Content 2008 • Health Insurance Coverage • Veteran’s Service-connected Disability • Marital History 2009 • Field of Undergraduate Degree • Wording and format changes in 2008 to match Census 2010 questions
2008 Content Changes Three new questions Health Insurance Coverage Veteran’s Service-connected Disability Marital History Deletion of one question Time and main reason for staying at the address Changes in some wording and format 16 16
ACS Sample Design • Sample is accumulated over TIME to produce lowest levels of geographic detail to replace census sample • 5 years of data are required for areas and population groups with less than 20,000 population • Sample cases selected from an updated Master Address File (MAF)
Sample • Questionnaires mailed to about 1 in 480 addresses each month throughout the U.S. and Puerto Rico. • 1 in 40 addresses per year (2.5% of the population). • Average of 500-600 addresses per month per congressional district. • Total of 3 million addresses each year, or 250,000 per month. • Inclusion of population in group quarters beginning in 2006.
Target Population • Resident population of the United States and Puerto Rico • Living in housing units and group quarters (group quarters added in 2006) • Current residents at the selected address
Methodology Sample includes about 3 million addresses each year Three modes of data collection mail phone personal visit Data are collected continuously throughout the year 20 20
Period Estimates Describe the average characteristics over a specific period of time Contrast with point-in-time estimates Do not describe the characteristics on a specific date Period is 1 year, 3 years, or 5 years for ACS
Geographic Boundaries • Multiyear estimates are based on geographic boundaries as of January 1 of the last year in the multiyear period Example: 2006-2008 ACS estimates use boundaries as of Jan 1, 2008 • Boundaries of other statistical areas will be updated every decade in conjunction with the decennial census • 2010 Census boundaries will be used for data released in 2011
ACS Data Release Schedule Hypothetical situation: If Census 2010 contained a long form, detailed characteristic data would not be available until 2012
ACS Data Products Profiles • Data Profiles • Narrative Profiles • Comparison Profiles • Selected Population Profiles Tables • Detailed Tables and Collapsed Tables • Subject Tables • Ranking Tables • Geographic Comparison Tables Thematic Maps Public Use Microdata Sample (PUMS) Files
Using the Data:Factors to Consider • Universe and residence rules • Time Periods • Reference Periods
Residence Rules • The American Community Survey uses a • “two-month” rule • Decennial census based upon the concept of • “usual residence”
Residence Rules • Resident of a housing unit if a person: - Lives there year round - Lives there more than 2 months but not year round - Is living there now with no other place to live - Is away now for 2 months or less • Not a resident of a housing unit if a • person: - Lives there 2 months or less with another residence - Is away now for more than 2 months
Period Estimates • Describes the characteristics of an area • over a specific period of time • Contrasts with point-in-time estimates • that describe the characteristics of an • area on a specific date • 1-year, 3-year, and 5-year estimates are released for geographic areas that meet specific population thresholds
Reference Periods ACS uses the interview date as the single reference point, or as the end of a reference period, for all data collection.
Questions With No Specific Reference Period • Most ACS questions • do not stipulate a • period of time that • should be referenced • Interpretation is yearly • average since the • data are collected • each month and • averaged across months
Questions With a Specific Reference Period Relative to Interview Date • Other questions specify a period of time relative to the date of interview • Interpretation is still a • yearly average but covering a slightly different period of time than the calendar year
Group Quarters • Place where people live or stay that is normally owned or managed by an entity or organization providing housing or services for the residents. Two types of group quarters: 1. Institutional 2. Non-institutional • Group Quarters Population includes all people not • living in households. -This term includes those people residing in group quarters as of the date the ACS was conducted.
Overview of ACS Timeline • First year of data collection for full sample in 2005. • Data for calendar year 2007 released beginning in August 2008. • Annual data for geographic areas over 65,000 population. • 3 year averages are now available for geographic areas 20,000 to 65,000. • 5 year averages for geographic areas under 20,000 in 2010.
Timeline • ACS Data single year collection (e.g. 2007) is closed out just after the beginning of a calendar year (e.g. 2008) • Single-year and multiyear data products start to become available in the summer of the same year. • For example 2007 ACS estimates were released in 2008 • The cycle repeats EVERY year
Estimates • ACS data are estimates • ACS data are not counts of the population or housing • Population counts are produced from the decennial census -Counts are updated throughout the decade through the Population Estimates Program
Margin of Error (MOE) • Margin of Error Definition: A measure of the precision of an estimate at a given level of confidence (90%, 95%, 99%) • MOEs at the 90% confidence level for all published ACS estimates • Confidence Interval Definition: A range that is expected to contain the population value of the characteristic with a known probability.
Interpreting Margin of Error • Indicates that a data user can be 90 percent certain that the estimate and the population value differ by no more than the value of the MOE • MOE can help data users assess the reliability of an estimate • MOE can help data users avoid misinterpreting small differences between estimates as significant
American Community Survey: Multiyear Data 2008 2006 2007
Constructing Multiyear Estimates • Data are pooled across 36 or 60 months • Data are weighted to produce estimates • Estimates are controlled for age, sex, race, and Hispanic origin • Multiyear estimates are not an average of 1-year estimates
Use Multiyear Estimates When … • No 1-year estimate is available • Margins of error for 1-year estimates are larger than required • Analyzing data for small population groups
Inflation Adjustment • Dollar-valued data items are inflation adjusted to the most recent year for the period • Income, rent, home value, and energy costs • Adjusted using inflation factors based on the Consumer Price Index (CPI) • Adjustment designed to put the data into dollars with equal purchasing power
Geographic Boundaries • Multiyear estimates are based on geographic boundaries as of January 1 of the last year in the multiyear period • Boundary Annexation Survey collects boundary changes • Boundaries of other statistical areas will be updated every decade in conjunction with the decennial census