Surveys are subject to two types of error, sampling error and nonsampling error. There is no sampling error because the Census of Governments covers all governments.  However, the data are subject to nonsampling error, which includes all other sources of survey error such as nonresponse, incorrect reporting, coverage, misclassification of governments, and inaccurate coding or processing of data.

A variety of procedures were applied to keep nonsampling errors to a minimum, including the following:

1. Review of state legislation for all Government Units Survey (GUS) responses that indicated that the entity was no longer a government 


2. Review of secondary sources


3. Reviews of lists of governments supplied by the states, when available


4. Nonresponse follow-up calls to encourage response


5. Extensive research on addresses for Undeliverable As Addressed (UAAs) units and other units lacking adequate contact information


6. Initial mail, reminder letters, and reminder emails for the GUS