OLMIS

by Gail Krumenauer
Published Dec-22-2010

How and Why We Count the Population New for the 2010 Census Getting the Census Results Temporary Employment Boost Supplemental Uses of Census Data

In 2010, the U.S. Census Bureau completed its decennial count of the population. The decennial census began in 1790, and continues as a constitutional mandate for apportionment purposes. Apportionment is the process of dividing the 435 seats in the House of Representatives among the 50 states to achieve proportionate representation of the population in the arena of public matters. Since 1975, the Census Bureau has also taken responsibility for creating sub-state population data used to define legislative and congressional districts.

Over the decades, the census has changed significantly. In the 1790 decennial count, the Census Bureau tells us that assistant marshals carried out the survey process by asking the number of "free white males, free white females, other free persons, and slaves" occupying each housing unit. By 1890, census enumerators - those conducting the population counts - were collecting the full name, race, sex, and age of each individual living in a household.

When I served as an enumerator during the 2000 Census, the population count was tallied using two different forms: a "short form" that included 8 basic questions, and the "long form" that was distributed to 1 of every 6 households nationwide. The census long form asked the same questions as the short form, along with 45 additional questions about everything from educational attainment and physical disabilities to employment status and income. The 2000 Census occurred mostly by mail, and enumerators only visited the households for non-response follow up if a form was not returned by mail.

The years between the 2000 and 2010 Census efforts have brought another substantial change to the process of counting our nation's population: the American Community Survey (ACS). The ACS currently publishes 1-year and 3-year estimates of population characteristics in equal or greater detail than the decennial census long form. In fact, the ACS has now replaced the long form entirely; all 2010 Census forms had only 10 questions.

With the American Community Survey handling the more detailed demographic questions, one might ask why we should still carry out the decennial census. The limitations of the American Community Survey are twofold. First, the survey covers less than 10 percent of the population, whereas the decennial census captures approximately three-fourths of the population. The scale of data collection in the decennial census provides far more accurate results. Second, the ACS results are only available for areas with a population above 65,000 for its 1-year estimates, and areas with a population exceeding 20,000 for the 3-year estimates. This leaves the decennial census as the sole source of valuable data for rural areas. In December 2010, the Census Bureau released the first 5-year ACS estimates for all areas. This removed the coverage limitation of the 1-year and 5-year ACS.

As of early December, the only census results published thus far include the state and county participation rates in the decennial festivities. Nationwide, the Census Bureau estimates a 74 percent participation rate in 2010, which matches the share of households participating in the 2000 count. Oregon has a higher response rate than the U.S. in each of the two most recent census efforts (76% in 2010 and 77% in 2000).

Within the state, census participation rates ranged from a high of 80 percent in Benton County to a low of 57 percent in Gilliam County (Figure 1). The highest rates of participation were concentrated in the Willamette Valley.

Soon, the data at the heart of the 2010 Census start to become available. The apportionment counts - which include the resident population of the 50 states, along with military and federal civilian employees and dependents who could be allocated to a state - are scheduled for release by December 31, 2010. The 2010 Census data planned release schedule shows that state population counts used in redistricting will become available between February and March of 2011. This release will also include data by race and Hispanic ethnicity. Between May and August, additional data releases will include more in-depth population counts by race and ethnicity, along with selected housing characteristics. Data associated with the 2010 Census will continue to be released into 2013.

The 2010 Census occurred during a sluggish economic time, one that continues today. As census employment and work began to ramp up in Oregon during January 2010, more than 225,000 Oregonians were unemployed. Census-related work gave a temporary employment bump, however small, that provided job opportunities to Oregonians who may not have otherwise had them.

The census employment patterns show a notable effect on Oregon's monthly employment numbers. Estimates generated by the Employment Department through the Current Employment Statistics (CES) program show that as census-related employment ramped up between January and May, the state's nonfarm payroll employment experienced either larger gains or fewer losses than Oregon would have otherwise experienced (Graph 1). As temporary employment for census workers declined between June and September, estimates show greater employment losses or fewer gains than would have occurred in an intercensal year.

May and June are the most illustrative months of this trend. In May, census employment hit its peak. Over-the-month employment declined by 1,800 jobs between April and May without counting census employment. With census work factored into the employment numbers, Oregon gained 2,600 jobs on a seasonally-adjusted basis, a difference of 4,400. In June, temporary census employment began to draw down sharply. Over the month, seasonally adjusted statewide employment increased by 4,400 jobs, but census layoffs brought monthly employment gains down to just 1,800, a difference of -2,600 jobs.

As the census process and content have expanded over time, so have the uses of the data collected. In addition to apportioning Congressional seats, the decennial census - and now also the American Community Survey - provide the most thorough demographic portrait of our society that's available.

When an employer calls us at the Oregon Employment Department and asks the size of the workforce ages 18 to 64 in Medford or Ontario, we look to census data for the answer. If you want to know the share of Oregonians with a 4-year college degree, and where Oregon ranks among all states, the census has your numbers. The information collected through the census and American Community Survey is used to serve businesses, nonprofit organizations, and public entities on a daily basis.

To learn more about decennial census data or the American Community Survey, visit http://factfinder.census.gov.

For information related to the Employment Department's CES program, visit the Current Employment Statistics page at www.qualityinfo.org/olmisj/CES.