E-6 County Population Estimates and Components of Change — July 1, 1990-2000

February 2005

Suggested Citation

State of California, Department of Finance, Revised County Population Estimates and
Components of Change by County, July 1, 1990-2000. Sacramento, California, February 2005.

This updated report presents revised state and county population estimates benchmarked to final 1990 and 2000 decennial census data. The statistical “error of closure” procedure corrects the estimation error in the original estimates. These estimates supersede those in the Department of Finance E-6 series for the 1990s published in January 2003.

Methodology

The state and county populations are independently estimated using population change models benchmarked on decennial census counts. The state population is estimated using the Driver License Address Change method. County population proportions are estimated using the average of three separately estimated sets of proportions as described below. The final distribution of proportions is applied to the independently estimated state control.

State. The state population is estimated using the Driver License Address Change (DLAC) method. This composite method takes account of annual births and deaths and features distinctive approaches to three age groups: persons under age 18; those aged 18 through 64; and those aged 65 and over. The estimate for the youngest group features changes in school enrollment and progression ratios, while that for the oldest group is based on changes in Medicare, Medi-Cal, and CHAMPVA enrollments. The estimate for the middle group relies on address changes recorded in the California Driver License Address Change file that have been adjusted using matched federal tax return data, and on immigration data from the U.S. Immigration and Naturalization Service. Survived cohorts of the household population in the age groups under 65 are adjusted by adding group-specific estimates of migration and group quarters population.

Counties. County population proportions are estimated using three methods whose results are averaged.

DLAC Method. A modified version of the Driver License Address Change method is used for counties, with shifts in the allocation of estimated proportions resulting from relative annual change in county population values as a function of births, deaths, and foreign and domestic migration.

Ratio-Correlation Method. This method models change in household population as a function of changes in the distributions of driver licenses, enrollments and labor force. Estimates of county group quarters are added.

Tax Return Method. County proportions are derived by the U.S. Census Bureau using matched federal income tax returns to estimate intercounty migration along with vital statistics, group quarters and other information for the population aged 65 and over.
Data Considerations

Sources. Data used in estimation models come from administrative records of 17 state and federal departments and agencies. Since timeliness and coverage in these series vary, corrections, smoothing and other adjustments are applied.

Accuracy. Data and models used to produce estimates are subject to measurement and nonmeasurement error, resulting in imperfect correlations between the data used and actual population change. However, these data adjustments and models have been thoroughly tested, and are further refined and modified through comparison with decennial census benchmarks. Results of the county estimation models used differed from decennial census counts by an average of 2 percent in 1990 and 1.9 percent in 2000. The total state estimate was within one-third of one percent (0.3%) of the decennial census count in 1990 and one-half of one percent (0.5%) in 2000.