sus questionnaire, to jog respondents’ memories and to provide further clues to census staff on the appropriate resolution of “usual residence.”
For the 1900 census—decades before self-response questionnaires, when the census was still conducted by enumerators recording entries in large, columnar ledgers—enumerators received a unique instruction. Concerning “inmates of hospitals or other institutions,” the instructions directed that all such inmates should be enumerated. However, the instruction concluded, “if they have some other permanent place of residence, write it in the margin of the schedule on the left-hand side of the page.” What, if anything, was done with any such information scrawled on the margins is unknown.
As described in Chapter 3, practice varied over subsequent decades as to whether such groups as hospital patients and military personnel stationed at domestic bases (but living in an off-base property) were to be counted at the institution or base or at another place. The spirit of the 1900 instruction and later experience ultimately led to the practice of “usual home elsewhere” (UHE) data collection. A major part of what limited organization exists in the formal residence rules of the 2000 census is the idea of reporting another address as a UHE, a privilege that the rules granted to some types of group quarters residents but not others. (However, as we will discuss in Chapter 7, the Individual Census Reports used for group quarters data collection asked for a UHE from all residents; only the information from members of permitted groups was further processed.) In recent censuses, UHE reporting has been solely limited to the group quarters population, with the exception of the provision in the 1970–1990 censuses of a checkbox for indicating that everyone at the address has a usual place of residence elsewhere and is only staying at the questionnaire address temporarily. Under those conditions only, the 1980–1990 versions asked for the full street address of the other place of residence.
The collection of accurate residence data is central to the census mandate, and the living situations in which people have legitimate, strong ties to more than one residence is not limited to the group quarters population. Collecting information on another place of residence, if applicable, for all persons on the census form better equips the Census Bureau to determine where each individual should be counted, as well as making the structure of the census form’s residence questions more consistent with real-life settings. The Census Bureau should strive to collect alternative residence (address) information from all census respondents, not a selected subset.
Recommendation 6.2: Information on “any residence elsewhere” (ARE) should be collected from census respondents. This in-