The CATI nonresponse followup phase of the Census Bureau’s 1997 test panel of the American Community Survey (ACS) included an experimental test of two instrument structures — the traditional “person-based” approach versus a new “topic-based” design. The former in essence completes an entire, separate interview for each eligible household member in turn; in contrast, a topic-based interview gathers data on one “topic” for every person and then proceeds to the next topic, in effect making only one “pass” through the instrument. Section 2 of this paper offers some basic background information about the research, including descriptions of the ACS, the characteristics of person-based and topic-based instrument designs, and the results of a small-scale pilot test of topic-based interviewing. In section 3 we summarize the design of the main experiment. Section 4 is devoted to the results of the experiment, focusing on four outcome domains: the interview process, interviewer and respondent assessments, and survey data effects. In section 5 we offer our summary and conclusions.