State and Local Government Employment and Payroll
March 2004
 
__
 Main State and Local Employment     2004 State and Local Employment 
__
Sample Design
    Description of design:
  The 2004 Sample for the Annual Employment Survey was developed in 2003 from the 2002 Census of Governments. It was designed to produce state by type of government estimates with a relative standard error of 3 percent or less on FTE employees and total payroll. Units satisfying the following criteria were automatically included in the sample with a probability of 1. These units represent themselves only.

               1.   All county governments with a 2002 population of 100,000 or more.
               2.   All municipalities with a 2002 population of 75,000 or more.
               3.   All townships with a 2002 population of 50,000 or more.
               4.   All independent school districts with an enrollment of 10,000 or more.
               5.   All school districts providing college level (postsecondary) education.
               6.   All special districts that meet at least one of the following criteria:
                       a.       FTE of 1,000 or more.
                       b.       In the state of Connecticut with a function code of 91 (Water Utility).
                       c.       In the state of Maine, New Hampshire, Rhode Island, Utah, or Wisconsin with a function code of 92 (Electric Utilities).
                       d.       Has a function code of 93 (Gas Utility).
  All other units were given a chance of coming into the sample based on the FTE employment and total payroll of the unit. Prior to selecting the sample, the file was ordered by state; within state, by type of government; and within type of government by population for cities, counties, and townships and by enrollment for school districts. For special districts, the file was ordered by probability of selection within function code.

Prior to mail-out, analysts update the sample universe file with births (units that did not exist during prior years of the survey) that have come into existence since the prior year processing cycle. All city, county, township, and school district births are added to the sample with a probability of selection of 1. Special district births are sampled at a rate of 1 in 25.

    Weighting:
  The weight for each unit in the sample is the inverse of that unit’s probability of being selected into the sample. For example, for units that were included in the sample with a probability of .02, the weight is (1/.02)=50. For units that were included in the sample with a probability of 1, the weight is 1.

    Sample size:
  The sample size is around 11,000 units. Of the total number of governments in the sample, approximately 13% are counties, 23% are cities, 14% are townships, 30% are special districts, and 19% are school districts. All 50 state governments and the District of Columbia are certainty units with a weight of 1.


__
 
Source: U.S. Census Bureau, Governments Division,
Created: 03-25-2005
Last revised: June 21 2005