1. Census Data-Strictly Business?:
What’s Different about Census 2010 (and on)?
What’s “New” late ?—
what’s in there and what’s not in there?
Access Points & Methods—demos
Q & A (ask at anytime)
Jim Skinner- Research Division
Atlanta Regional Commission

2. Decennial Census in the recent past…
In Census 2000, the census 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 (SF1 and SF2)
“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 (SF3..and SF4)
Only a small portion of the population, called a sample, received the long form. 1 IN 6 IN 2000
Two forms were used during Census The “short form” asked several questions on the most basic demographic and housing topics. These questions asked about age, sex, race, Hispanic origin, the number of people living in the housing unit, and if the housing unit was owned or rented by the resident.
The “long form” collected the same information as the “short form,” but it also asked questions on additional topics. Thirty-two questions were asked of each resident of the housing unit on such topics as marital status, education, language spoken at home, employment, occupation, and others. Twenty-one questions were asked about the housing unit itself, so only one resident of that housing unit was asked to provide information on such topics as plumbing and kitchen facilities, type and cost of utilities, value of the property, and others.
Only a sample of the total U.S. population received the long form. The data from the long form are called “sample data.” The basic data collected on both the short and the long forms are called “100 percent data” since these questions were asked for 100 percent of the U.S. population. 2 2

3. 2010 Census and American Community Survey
2010 Census focused on counting the U.S. population
The sample data are now collected in the ACS. There is no more “long-form” (no more SF3 or SF4)
Same “short form” questions on the ACS
5-year estimates will be produced for same broad set of geographic areas including census tracts and block groups
The upcoming 2010 Census will include only one form sent to the entire U.S. population. That form will ask only questions similar to those contained in previous census short forms. The 2010 Census will provide a basic count of the U.S. population, collecting only the most basic demographic and housing information. Detailed demographic, social, economic, and housing data will no longer be collected as part of the decennial census.
The data that were collected from the long form sample are now produced from the American Community Survey.
The American Community Survey collects data from the 50 states, the District of Columbia, and Puerto Rico, where it is called the Puerto Rico Community Survey. The American Community Survey does not collect data from the other U.S. territories, such as Guam and the U.S. Virgin Islands. The 2010 decennial census will continue to collect long form type data for these areas.
The questions that are asked on the 2010 Census are also asked on the American Community Survey questionnaire. 3 3

4. American Community Survey Key Differences from Census 2000
The goal of ACS is to produce data comparable to the Census 2000 long form data
These estimates will cover the same small areas as Census 2000 but with smaller sample sizes
Smaller sample sizes for 5-year ACS estimates results in reductions in the reliability of estimates
In Census 2000 approximately 17 percent of the housing units received a long form. Statistics from this sample were produced for a broad set of geographic areas including the nation, all states, counties, census tracts, and block groups.
Five years of American Community Survey data are needed to produce estimates comparable to the estimates produced from the Census 2000 long form. A benefit that users will gain from the American Community Survey is the more timely issuance of the data and the greater frequency with which the data are released. Also, it produces information for small areas, including tracts and block groups, which will be updated every year instead of once every decade.
The sample sizes for the 5-year ACS estimates will be smaller than the sample sizes were for Census This will result in a reduction in the reliability of the estimate. 4 4

5. Population Thresholds for ACS Estimates
1-year estimates
3-year estimates
5-year estimates
65,000 + people X
20,000+ people
Less than 20,000 people
The types of ACS estimates published for a particular area are based on established population thresholds.
Geographic areas with at least 65,000 people will receive one-year, three-year, and five-year ACS estimates.
Areas with 20,000 or more people will receive three-year and five-year estimates. There are a few exceptions to this rule, however. Zip code tabulation areas, census tracts, and block groups, regardless of their population size, will only receive five-year estimates.
Areas with less than 20,000 people, down to the block group level, will only receive five-year estimates. 5

6. American Community Survey Data Products Release Schedule
Population Size
Data rele
ased in:
of Area
2006
2007
2008
2009
2010
2011
2012
2013
1-Year Estimates
65,000+
2005
for Data Collected in:
3-Year Estimates
20,000+
5-Year Estimates
All Areas*
* Five-year estimates will be available for areas as small as census tracts and block groups.
Source: US Census Bureau
This slide shows the data products release schedule for 2006 through 2013.
All American Community Survey data are published based on population thresholds. One-year estimates from the American Community Survey are published for geographic areas with a population of 65,000 or more, including counties or county equivalents, all Congressional Districts, metropolitan and micropolitan statistical areas, all 50 states, the District of Columbia, and Puerto Rico American Community Survey one-year estimates were released in September 2008.
Data will be available for areas with estimated populations of 20,000 or greater in December These three-year estimates will be produced with data collected from the 2005, 2006 and 2007 American Community Surveys.
The Census Bureau plans to release data for all areas, including those with estimated populations of less than 20,000, by the end of For small areas with populations less than 20,000, it will take five years to accumulate a large enough sample to provide estimates with accuracy similar to the decennial census long form. These five-year estimates will be produced from the data collected from the 2005, 2006, 2007, 2008, and 2009 American Community Surveys.
Once published for the first time, all American Community Survey data products are refreshed annually. 6 6

7. CENSUS DATA: OVERVIEW OF The laST YEAR+
2010 Census Redistricting Data
Counts released for Georgia 3/17/11
State, Counties, Cities, Census Tracts, Blockgroups, Blocks
Total Population by Race (Detail), and Voting Age, Housing Units
NO Age or gender detail, NO income and poverty
American Community Survey Data
Estimates released for Georgia 12/2010
State, Counties, Cities, Census Tracts, Blockgroups
Everything in above release (but not as current) plus age and sex, income, poverty, mode choice, commuting, etc.

8. CENSUS DATA: OVERVIEW OF The laST YEAR+ (CONTINUED)
2010 Census SF1 Data
Counts released for Georgia June 2011
State, Counties, Cities, Census Tracts, Blockgroups, Blocks
Total Population by Race (Detail), Sex and Age Detail, More Detail on Housing Units
STILL, no income and poverty data
American Community Survey Data
Estimates to be released for Georgia 12/2011
State, Counties, Cities, Census Tracts, Blockgroups
Everything in above release (but not as current) plus age and sex, income, poverty, mode choice, commuting, etc.
2010 American Community Survey (for similar variables)
released January 2012
LARGER geographies (65K +); ACS down to 20K+ pop
NOTE: No “sample” data from 2010 Census (no SF3)…not until 2013 for ACS

9. WHAT’S OUT AND IN THE NEWS…
2010 Census SF2
Counts released for Georgia late March 2012
State, Counties, Cities, Census Tracts
331 racial and ethnic categories: Total Population, Sex and Age Detail, More Detail on Group Quarters
VERY complicated table structure
2010 Urbanized Areas
2011 Census Bureau Population Estimates
Estimates released for Georgia 4/2012
State, Counties, Metro and Micro Areas (“OLD” definitions)

11. Urban Area by County: 2010

12. 2011 Bureau Estimates Compared

13. CENSUS DATA: OVERVIEW OF The laST YEAR+ (CONTINUED)
2010 Census SF1 Data
Counts released for Georgia June 2011
State, Counties, Cities, Census Tracts, Blockgroups, Blocks
Total Population by Race (Detail), Sex and Age Detail, More Detail on Housing Units
STILL, no income and poverty data
American Community Survey Data
Estimates to be released for Georgia 12/2011
State, Counties, Cities, Census Tracts, Blockgroups
Everything in above release (but not as current) plus age and sex, income, poverty, mode choice, commuting, etc.
2010 American Community Survey (for similar variables)
released January 2012
LARGER geographies (65K +); ACS down to 20K+ pop
NOTE: No “sample” data from 2010 Census (no SF3)…not until 2013 for ACS

15. Detailed Tables Examples
Sex by Age by Race and Hispanic Origin
Means of Transportation to Work by Travel Time to Work
Median Number of Rooms in Housing Units
School Enrollment by Level of School
Poverty Status in the past 12 Months by Sex and Age
The detailed tables provide the most detailed data on all topics and geographic areas and are the foundation upon which other data products are built.
Detailed Tables are basic distributions of characteristics that show estimates and their associated margins of error. There are more than 1,200 Detailed Tables and many of the tables are repeated for 11 race and Hispanic origin groups. Detailed tables include distributions for more than 500 characteristics, over 300 race and Hispanic Origin iterations, and 81 imputation tables.
A few examples of the types of topics covered by our detailed tables…
Sex by Age by Race and Hispanic Origin
Means of Transportation to Work by Travel Time to Work
Median Number of Rooms in Housing Units
School Enrollment by Level of School
Poverty Status in the past 12 Months by Sex and Age 15 15

16. Detailed Tables Table Name Table Number Universe Data Source
Collapsed Version
Margin of Error
This slide displays an example of a detailed table as displayed on the American FactFinder. The table is B Age by Language Spoken at Home for the Population 5 Years and Over from the 2007 American Community Survey 1-Year Estimates.
The table title at the top describes the variables in the table, and any combination of them for which estimates are presented and the universe for which the estimates apply. The data set indicates the year the data were collected and the type of estimate.
The table has an additional column containing the margin of error for the 90 percent confidence level of the estimate. Confidence bounds can be created by adding the margin of error to the estimate (for an upper bound) and subtracting the margin of error from the estimate (for a lower bound). All published margins of error for the American Community Survey are based on a 90 percent confidence level.
Due to the level of specificity of the detailed tables, many tables have a “collapsed” version. If a full detailed table has a collapsed version, it will be indicated at the top of the table. 16 16

17. American Community Survey

30. LIVE DEMOS AFF2 DEMO ARC CENSUS SITE WEAVE INTERFACE (via ARC) >>NEIGHBORHOOD NEXUS OTHER (IRE, NHGIS)