Does a Final Coverage Check Identify Census Omissions and Reduce Errors?
Preliminary Results of an Experiment
Elizabeth Martin and Don A. Dillman
KEY WORDS: questionnaire design, coverage errors, decennial census, split-panel experiment
In 2010, automated matching of name and date of birth for every census enumeration will better identify duplicates than in past censuses, but the problem of omissions persists. Ironically, eliminating duplicates might result in a larger net census undercount. This research evaluates new questions to identify and reduce omissions and other coverage errors. At the end of the short form, respondents are given reminders of people who might be missed, and asked to review the form and answer two questions about possible errors.
In a March 2006 national test, experimental and control forms were mailed to random samples of households. Results indicate that most respondents found and answered two final questions intended to identify coverage errors. The situations described in their open-ended answers include many that are known to give rise to census coverage errors.
When followed up with a coverage followup (CFU) interview, 4.5% of the households flagged by the Final Questions added a person to the roster, compared to 0.7% of those not flagged by the FQ. Thus, these questions do help discriminate between households in which a follow up interview is productive from those in which it is much less so.
The rate of additions seems low in relation to a relatively high proportion of write-in responses that described potential residence situations. In part, CFU obtained more detailed information that determined many cases to be residents of other places. In addition, CFU failed to identify as possible adds many of the possible missed residents described in write-in responses.
CITATION: Elizabeth Martin and Don A. Dillman. 2006. “Does a Final Coverage Check Identify Census Omissions and Reduce Errors? Preliminary Results of an Experiment.” 2006 Proceedings of the American Statistical Association, Survey Research Methods Section. [CD-ROM] Alexandria VA: American Statistical Association.