Top-Coding and Public Use Microdata Samples from the U.S. Census Bureau

The US Census Bureau regularly releases Public Use Microdata Samples (PUMS), data files which contain de-identified subsets of the data provided by respondents to some of its various surveys and to the Decennial Census itself. This allows data users to perform “micro” -analyses rather than the “macro” -tabulations which are regularly performed by the Bureau. These data users range from non-government (say, university) researchers to government policymakers. These micro-analyses typically depend on the joint distribution of two or more variables over individuals or households. As a very simple example, think of the relationship of wages of individuals to their individual ages by a linear regression equation. We will use this very simple example throughout this paper to illustrate the effects we are interested in. In order to protect the privacy of the data supplied by respondents, as required by Title 13 U.S.C., the Bureau uses a variety of methods to modify the data so that it is very difficult for data users to identify individual respondents. Although some kind of privacy protection measures are necessary by law, most of them (top-coding, in particular) have a detrimental effect on the micro-analyses because application of these privacy protection measures changes the interdependence of two or more variables and, in many cases, renders the analyses moot.