Cognitive Testing Plan -- 2010 Alternative Questionnaire Experiment (Overcount Questions)

ombCPEXovercount0903.doc

Generic Clearence for Questionnaire Pretesting Research

Cognitive Testing Plan -- 2010 Alternative Questionnaire Experiment (Overcount Questions)

OMB: 0607-0725

Document [doc]
Download: doc | pdf

The Census Bureau plans to conduct cognitive testing of a new series of overcount questions designed for use in a coverage panel (designed in booklet form) in the 2010 Census Alternative Questionnaire Experiment. This is part of the 2010 Census Program for Evaluations and Experiments.


The control version of the 2010 census form includes one overcount question, “Does Person 1 sometimes live or stay somewhere else?” This question has no/yes checkoff boxes, then eight response categories that identify the type of place or situation for those who marked “yes”: “in college housing”; “in the military”; “at a seasonal or second residence;” “for child custody;” “in jail or prison;” “in a nursing home;” and “for another reason.” The question serves as a flag to identify households with possible erroneous enumerations for later inclusion in the Coverage Followup Operation to determine where the persons with other places should be counted in the census.


The experimental version we will be testing asks a series of questions to try to determine usual residence from answers on the form itself, thereby avoiding coverage followup for some types of living situations. The draft version of these questions on a mocked-up person page of the questionnaire booklet is attached (the same sequence of questions will be asked for Persons 1 to 9 in the final version of the booklet). This battery of questions starts with the same question stem above, “Does Person 1 sometimes live or stay somewhere else?” with the no/yes checkoff boxes, but includes the following changes: a reversal in the order of presentation of the “for child custody” and “at a seasonal or second residence” response categories, and the insertion of a new category, “for a job or business.” This latter category has been included in previous tests of the overcount question over the decade but was removed due to space limitations on the control form.


For those who mark “yes” to this screening question, the experimental version also includes a request for the full address of the other place where Person 1 (and subsequent persons) sometimes live or stay; a request for a write-in description if this place has no street address or if this is a facility; and two residence rule-related questions: “Where does Person 1 live or stay most of the time?” and “On April 1, where was Person 1 staying?” These two questions capture de jure and de facto residence information that may be used during processing of the census form to determine the person’s residence status in some (but not all) situations.


If this sequence of questions works as intended to determine residence status for some proportion of the population from the questionnaire itself, the number of households needing Coverage Followup would be reduced, saving time and labor and reducing costs.


The cognitive interviews will commence in November and conclude in December. We plan to conduct 15 to 20 cognitive interviews in the Washington, DC metropolitan area in our lab or at a place more convenient for respondents. In order to get a good test of these questions, we will selectively recruit respondents likely to have one or more persons in their households who sometimes live or stay somewhere else, both in the types of living situations identified in the lead-in question response categories and in other types of mobile situations. We will recruit people of different races and origins, genders, and education and income levels through use of our database, personal connections, and craigslist.org.


Prospective respondents will be informed that their participation is voluntary and that information that could identify individuals will be held strictly confidential. The information will be used only for research purposes. Interviews will be audiotaped with the permission of the respondents. Respondents will be paid $40 as an honorarium.


The estimated time for completion of the interview is one hour. Thus, the maximum estimated burden for this activity is 20 hours.


The contact person for questions regarding this data collection is:


Laurie Schwede

Center for Survey Methods Research

Statistical Research Division

U.S. Census Bureau

4600 Silver Hill Road

Washington, D.C. 20233

(301) 763-2611

Laurel.K.Schwede@census.gov


File Typeapplication/msword
File TitleDear Brian,
AuthorBureau Of The Census
Last Modified Bysmith056
File Modified2008-10-23
File Created2008-10-23

© 2024 OMB.report | Privacy Policy