Skip Main Navigation Skip To Navigation Content

Research Reports

You are here: Census.govSubjects A to ZResearch Reports Sorted by Year › Abstract of SSM2007/16
Skip top of page navigation

BEHAVIOR CODING ANALYSIS REPORT: Evaluating the English and the Spanish Versions of the Non-Response Follow-Up (NRFU) for the 2006 Census Test

Jennifer Hunter Childs, Ashley Landreth, Patricia Goerman, Dawn Norris, and Aref Dajani

KEY WORDS: Nonresponse Follow-up, behavior coding, pretesting, decennial, demographic questions, topic-based


Staff from the Statistical Research Division conducted an evaluation of the Nonresponse Followup (NRFU) automated instrument used in the 2006 Census Test conducted in Austin, Texas. Audiotapes of 101 NRFU interviews conducted in English and in Spanish were made during the data collection. Of these, a total of 72 usable interviews (54 in English and 18 in Spanish) were subsequently behavior coded by bilingual interviewing staff at the Tucson Telephone Center. Results from this study revealed that enumerators’ behavior was far from perfect. High rates of poor interviewer behavior may have changed the meaning for all but two questions (one on “relationship” and one on “age”). Across all questions, “ideal” interviewer behavior (i.e., asking the question as worded or with minor changes that would not affect meaning) occurred less than half of the time (40%). Interviewers found it particularly difficult to ask the questions on residence rules and coverage as well as the question on ownership as-worded. These results and the corresponding recommendations coincide with those from previous cognitive testing of these questions. At first glance, the results suggest the topic-based question administration approach did not result in more frequent reading of questions as-worded than did the person-based approach used in the NRFU 2004 instrument. Further analysis showed that this nature of the topic-based design may have affected this outcome. Interviewers were instructed to reread the question for each person in the household, one after another. However, if the instrument had allowed interviewers to use more conversational phrasing for asking the same question of subsequent household members (for example, “and Person 2/John?”), the topic-based approach would, in fact, increase the frequency with which interviewers read questions as intended. The report also contains a question-by-question evaluation of the questionnaire items, and includes recommendations for revising them.


Source: U.S. Census Bureau, Statistical Research Division

Created: June 14, 2007
Last revised: June 14, 2007

[PDF] or PDF denotes a file in Adobe’s Portable Document Format. To view the file, you will need the Adobe® Reader® Off Site available free from Adobe.

This symbol Off Site indicates a link to a non-government web site. Our linking to these sites does not constitute an endorsement of any products, services or the information found on them. Once you link to another site you are subject to the policies of the new site.

Source: U.S. Census Bureau | Statistical Research Division | (301) 763-3215 (or |   Last Revised: October 08, 2010