In recent decades, U.S. censuses have produced relatively accurate population counts, but the considerable importance of the results has driven efforts to study and possibly correct the errors of coverage that occur. In this paper, we analyze the recently available responses to individual questionnaire items designed to measure where people lived on Census Day, April 1, 2000. Our objective is to identify and clarify particular sources of error and bias, and to provide a basis for improving the design of future coverage measurement instruments. We introduce independent, auxiliary evidence of census duplications, available from a special study conducted to evaluate the quality of coverage estimates. We focus here on measurements of mobility; a more complete paper (available from the first author) includes additional analyses of other coverage measurements. We first briefly review sources of coverage error and the Census Bureau’s coverage measurement methods and describe the sources of our data.