• Visual cues should be used to convey what respondents are supposed to do. For example, white space (against a light background color) or some other graphical feature should be used to indicate when a response is required. Other graphical cues should indicate what form the response should take (e.g., boxes invite respondents to check one or more items).


As prologue to suggestions on how best to present residence concepts to respondents, it is instructive to consider the ways in which recent U.S. censuses have presented residence instructions and cues. We also discuss the approaches taken in census tests and follow-up census operations, as well as strategies used in foreign censuses.

Previous U.S. Censuses

As noted in Box 5-5, the 1950 census enumerator instructions were the first to include a detailed itemization of persons to include or exclude from the census. That approach clearly carried over to the 1960 instructions to respondents on the “Advance Census Report” mailed to households (Figure 6-1). The “PLEASE BE SURE TO LIST” and “DO NOT LIST” categories dominate the instructions at the top of the form, and overwhelm the basic “usual residence” statements at far left under “PLEASE LIST.” Other features of note in the 1960 instructions for respondents include the unconditional plea about “including babies,” the emphasis on people staying at the household but “who have no other home” (mentioned twice), and the emphasized assurance that college students, military personnel, and persons in institutions would be counted at their other location.

In preparation for the 1970 census—the first to be conducted principally by mailout/mailback—the Census Bureau developed “approximately 700 different questionnaires, field and administrative forms[,] address registers, handbooks, and manuals” for testing between 1961 and 1967. Experiments covered “type styles and sizes, paper and ink colors, as well as [the] formats in which various items would be printed” (U.S. Bureau of the Census, 1970:4-8). The final version of the 1970 household questionnaire included its residence instructions (Figure 6-2) on a separate page immediately preceding the listing of household members in Question 1. The number of specific include/exclude categories is higher than the number used in 1960,3 though at least one “addition” comes from listing college students twice—to list those living at home


The Bureau’s precedural history of the 1970 census suggests that, “if the enumerator found these instructions insufficient, he referred to a table of residence rules” for additional detail (U.S. Bureau of the Census, 1970:15-4).

The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement