California and Counties Population by Age, Race/Hispanics, and Gender: 2000–2010
September 6, 2012 (Original Release – integer data only)
September 14, 2012 (Revised to include decimal detail)
March 19, 2013 (Removed data for July 2010)
- Excel Data Files – one row for each Race/Hispanics and Age combination with Years in columns:
- Complete File Database Ready Format – Zipped file of comma-delimited text file and documentation.
Important Note: This very large comma-separated file contains 1,001,112 records and is intended to be imported into a database. It will NOT successfully open in Microsoft Excel.
These data files provide intercensal population estimates of single years of age, race and Hispanics, and gender for California and its counties. These estimates are for the time periods of April 1, 2000 and 2010 and July 1, 2000 through 2010.
The California Department of Public Health provided the vital statistics (births and deaths) used to develop these estimates. The Department of Finance’s own intercensal estimates of total county and state population correspond to the E-2 Report: California County Population Estimates and Components of Change by Year, July 1, 2000-2010.
The U.S. Census Bureau provided the 2000 and 2010 Census Modified Race files (www.census.gov). The total population by geographic area matches previously published counts in the 2000 and 2010 Census; however, the age, race, and gender details in the modified counts do not match published data. (See the Technical Notes and Assumptions below)
Intercensal population estimates are estimates made for the years between two completed censuses which take into account the census at both the beginning and end of the decade.
Basic Method: The Department of Finance used a baseline cohort-component method to estimate population by age, gender and race/Hispanics. A cohort-component method traces people starting from a given year through to the ending year. As each year passes, cohorts change as specified by actual deaths and migration assumptions. New cohorts are formed by applying actual births to the series every year and surviving them. For this series, there are seven mutually exclusive groups: Hispanics and six non-Hispanic race groups of Whites, Blacks, American Indians, Asians, Pacific Islanders, and Multiple-Race.
The intercensal estimates for 2000-2010 were generated by first producing forward and reverse cohort-component estimates. Forward component estimates were produced by updating the resident population enumerated in the 2000 census by the components of population change between April 1, 2000 and April 1, 2010. The components include births, deaths and net migration. Likewise reverse component estimates were produced by backdating the resident enumerated Census 2010 population to April 1, 2000, by adding deaths and net migration back into the series.
Yearly Age, Gender, and Race/Hispanics estimates were produced both for the resident population and for net migrants. These were controlled using county population and net migration totals taken from the independently produced E-2 report using the plus-minus method to ensure distributional consistency. The state is an additive sum of all the counties and was not independently controlled.
The forward and reverse cohort-component estimates of the race, gender and age distribution were then combined using a modification of the Das Gupta method, which assumes that the ratio of the intercensal estimate to the censal estimate should follow a geometric progression over the decade:
Pt = Ft (u / 120) + Rt (s/ 120)
s = time in months elapsed since April 1, 2000
u = time in months until April 1, 2010
Pt = Population estimate at time t
Ft = Forward Component estimate at time t
Rt = Reverse Component estimate at time t
Special Populations: Special populations are populations that do not age normally. They are found primarily in prisons, college dorms, and group housing on military installations. These populations tend to remain static in age as people enter and leave the institutions. In counties where special populations represent a significant proportion of a specific race/ethnic group’s population, they were removed from the base and handled separately.
Base Population: As the benchmark (the starting and ending population), the Department of Finance used respectively the 2000 and 2010 Census counts as modified by the Bureau of the Census to eliminate the “Other” race category. These counts, from the Modified Race files, represent a modification to the race distribution of the census count and not an adjustment to the total.
Migration: Migration proportions were developed for the decade of the 2000s by a survived population method. The 2000 population was aged forward in time to 2010 by adding recorded births to form new cohorts and subtracting deaths from existing cohorts. The survived 2000 population was compared to the Census 2010 population and differences were assumed to be migration. The ten-year migration was annualized and divided by the total to derive a yearly migration proportion. These proportions were then smoothed using a 3-point rectangular smoothing function. A similar method was used to produce the reverse cohort migration proportion.
These population estimates were prepared under the mandate of Government Code, Sections 13073 and 13073.5. In addition, the State Administrative Manual, Section 1100 on state plans, sets the general policy of . . . “(3) The use of the same population projections and demographic data that is provided by the State’s Demographic Research Unit.”
Walter Schwarm prepared these estimates of population by age, gender and race/Hispanics. Technical and analytical expertise was provided by Evaon Schnagl and Nancy Gemignani.
State of California, Department of Finance, Race/Hispanics Population with Age and Gender Detail, 2000–2010. Sacramento, California, September 2012.