Some survey questions do not apply to all respondents. How to design these questions for both eligible and ineligible respondents is unclear. This article compares full filter (FF) and quasi filter (QF) designs for autobiographical questions in mail surveys. Using data from National Health, Well-being, and Perspectives Study, we examine the effect of type of filter on item nonresponse rates, response errors, and response distributions. We find that QF questions are more confusing to respondents, resulting in higher rates of item nonresponse and response errors than FF questions. Additionally, FF questions more successfully identify ineligible respondents, bringing estimates closer to national benchmarks. We recommend that survey designers use FF designs rather than QF designs for autobiographical questions in mail surveys.
Get full access to this article
View all access options for this article.
References
1.
BogenK.1996. The effect of questionnaire length on response rates—A review of the literature. Paper presented at the Annual Conference of the American Association for Public Opinion Research, Salt Lake City, UT. https://ww2.amstat.org/sections/srms/Proceedings/papers/1996_177.pdf (accessed July 22, 2009).
DillmanD. A.SmythJ. D.ChristianL. M.. 2014. Internet, phone, mail, and mixed mode surveys: The tailored design method. Hoboken, NJ: Wiley.
4.
EckmanS.KreuterF.KirchnerA.JackleA.TourangeauR.PresserS.. 2014. Assessing the mechanisms of misreporting to filter questions in surveys. Public Opinion Quarterly78:721–33.
5.
GendallP.RamsayR.. 2001. An experiment with skip instructions. Marketing Bulletin12:1–9.
6.
KnauperB.1998. Filter questions and question interpretation: Presuppositions at work. Public Opinion Quarterly62:70–78.
7.
KnauperB.1999. The impact of age and education on response order effects in attitude measurement. Public Opinion Quarterly63:347–70.
8.
KnauperB.TurnerP. A.. 2003. Measuring health: Improving the validity of health assessments. Quality of Life Research12:81–89.
9.
KreuterF.McCullochS.PresserS.TourangeauR.. 2011. The effects of asking filter questions in interleafed versus grouped format. Sociological Methods & Research40:88–104.
10.
KrosnickJ. A.HolbrookA. L.BerentM. K.CarsonR. T.HanemannW. M.KoppR. J.MitchellR. C.PresserS.RuddP. A.SmithK.MoodyW. R.GreenM. C.ConawayM.. 2002. The impact of “no opinion” response options on data quality. Public Opinion Quarterly66:371–403.
NarayanS.KrosnickJ. A.. 1996. Education moderates some response effects in attitude measurement. Public Opinion Quarterly60:58–88.
16.
PoeG. S.SeemanI.McLaughlinJ.MehlE.DietzM.. 1988. “Don’t know” boxes in factual questions in a mail questionnaire: Effects on level and quality of response. Public Opinion Quarterly52:212–22.
17.
RedlineC. D.DillmanD. A.. 2002. The influence of alternative visual designs on respondents’ performance with branching instructions in self-administered questionnaires. In Survey nonresponse, eds. GrovesR.DillmanD.EltingeJ.LittleR., 179–96. New York: Wiley.
18.
SchaefferN. C.PresserS.. 2003. The science of asking questions. Annual Review of Sociology29:65–88.
19.
TourangeauR.CouperM. P.ConradF.. 2004. Spacing, position, and order: Interpretive heuristics for visual features of survey questions. Public Opinion Quarterly68:368–93.
20.
TurnerC. F.LesslerJ. T.GfroererJ. C., eds. 1992. Survey measurement of drug use: Methodological studies. Rockville, MD: National Institute on Drug Abuse, U.S. Department of Health and Human Services.
21.
U.S. Census Bureau. 2015a. Current population survey, Table H2. Households, by type, age of members, region of residence, and age of householder. http://www.census.gov/hhes/families/data/cps2015AVG.html (accessed February 18, 2016).
22.
U.S. Census Bureau. 2015b. Table AVG2. Average number of people per family household, by race and Hispanic origin, marital status, age, and education of householder. http://www.census.gov/hhes/families/data/cps2015AVG.html (accessed February 18, 2016).
