2. Component Method II. This method employs vital statistics to measure natural increase and school enrollment to measure net migration. The estimates made by the Census Bureau's Component Method II are specific to the civilian population under 65. To this population is added an estimate of the resident military population based on station strength statistics and an estimate of the population 65 and over based on Medicare statistics.1

3. The Composite method. In this method estimates of various age groups are derived separately and are then summed to secure a total for all ages. Death statistics are used to estimate the population in ages 45 and over, birth statistics are used to estimate the population in ages 18 to 44, and a variation of Component Method II is used to estimate the population in ages under 18. In the case of birth and death statistics, an arithmetic average of two years of data is used for both the base year (1970) and the estimate year (1972). The two-year average is used to smooth out random fluctuations in the data which distort the estimates.

1

inventory. Changes in the housing inventory are derived from data on building permits issued and demolition records, or on data on electric meter connections. The provisional July 1, 1973 estimates for the remaining counties were developed by adding the change between 1972 and 1973 Component Method II estimates to the 1972 estimates. All counties were subsequently adjusted to agree with the provisional July 1, 1973 State estimate published in Current Population Reports, Series P-25, No. 508.

Table 2 of this report presents estimates of the population of metropolitan areas and metropolitan counties in the State. The titles and definitions of the standard metropolitan statistical areas (SMSA's) are those currently defined by the Office of Management and Budget, Executive Office of the President. In New England, where towns rather than counties are the components of SMSA's, estimates are shown for the nearest equivalent metropolitan State Economic Area (SEA) or county.

Corresponding estimates for other States in the program will be published as they become available. The appendix table shows reports published to date for States in the 1972-73 series, together with those published earlier for 1971 and provisional 1972.

The estimates presented in the table have been rounded to the nearest hundred without being adjusted to the State total, which was independently rounded to the nearest thousand. Percentages are based on unrounded numbers.

Series P-26, No. 66

This report presents population estimates for July 1, 1972 and provisional estimates for July 1, 1973, for counties and metropolitan areas prepared under the auspices of the Federal-State Cooperative Program for Local Population Estimates. The objective of this program is the development and publication of State-prepared estimates of the population of counties using uniform procedures largely standardized for data. input and methodology. The methods used have been mutually agreed upon by the individual States and the Bureau of the Census on the basis of a test of methods against the 1970 census. For a more detailed description of the program and an analysis of the test results, see Current Population Reports, Series P-26, No. 21, "Federal-State Cooperative Program for Local Population Estimates: Test Results--April 1, 1970, April 1973.

The estimates shown here for the 39 counties in the State were prepared by the Population Studies Division of the Washington Office of Program Planning and Fiscal Management in cooperation with the Bureau of the Census. This agency was designated by the Governor to work with the Bureau of the Census in implementing and carrying out the Federal-State Cooperative Program.

The estimates shown for July 1, 1972 and July 1, 1973 are based on an average of the following methods, adjusted to agree with the July 1, 1972 and July 1, 1973 State estimates published in Series P-25, No. 508.

7

Issued May 1974

1. The Regression (ratio-correlation) method. In the Regression method a multiple regression equation is used to relate changes in a number of different data serie change in population distribution. The serie of data used in the Regression method for Washington are registered automobiles (X1), elementary school enrollment in grades 1 through 8 plus elementary special and elementary ungraded (X) and covered employment (X3). The prediction equation for Washington for the 1970's is given by