Geodemographic data resources, contd.
As was mentioned above, primary data consist of ad hoc surveys specifically measuring the subject matter of interest to the researcher and are generally limited to small samples taken at random across the nation such as political polls or consumer panels. In general, such primary surveys rarely can be used to infer facts about individual small geographic areas.

Secondary data, however, emanate from hundreds of millions of administrative records such as tax returns or from survey documents such as the decennial census schedules completed (in 1990) by all ninety-one million households in the United States or a substantial sample of them (one out of six). The geographic comprehensiveness and depth of detail of secondary sources make it possible to add such data up into accurate aggregate counts or statistics describing very small geographic areas and to make these results accessible to the public in "summary files" on CD ROM's, magnetic tapes, publications and online internet downloads.

For the most part, data from secondary sources are available for individuals rather than geographic areas only if their location is given so broadly that it would be impossible to identify them without a complete name and address (which, of course, is withheld by law). The privacy of each respondent, therefore, is in no way threatened by the dissemination of personal data. An example of such a dataset is the Public Use Microdata Sample of the Census of Population and Housing (PUMS). The location of any one individual in this file is given only in terms of areas containing no fewer than 100,000 persons as noted above.

More about geodemographic data resources...