Notice: the WebSM website has not been updated since the beginning of 2018.

Web Survey Bibliography

Title An Examination of How Survey Mode Affect Eligibility, Response and Health Condition Reporting Rates in Household Surveys and Whether Length of the Questionnaire Mitigates These Results?
Year 2016
Access date 09.06.2016
Due to decreasing rates of response for telephone surveys, a number of large, national level surveys have investigated or begun the transition to self-administered survey modes such as web and mail using address-based samples (ABS). However, the bulk of the research on mode effects is lacking in three ways. First, there are many studies that fail to use a true experimental design and instead compare results across a mixed mode survey to examine effects. Second, when there is an adequate design, the surveys tend to focus on cross-sectional local or student surveys, not national level results. Third, even when the two above requirements for accurate assessment are met, the effect of a screener for eligibility, sensitive questions, and the effects of questionnaire length are absent. To address this gap in the literature, we conducted a national level experiment using the National Survey of Children’s Health (NSCH). The NSCH project involved a redesign of the historically telephone survey that incorporates two significant design changes over prior administrations: it moved from a random-digit-dial (RDD) to ABS frame and it moved from interviewer-administered telephone data collection to self-administered mail and web data collection. In an effort to anticipate changes in estimates due to mode effects, a mode effects experiment was designed and conducted where in sampled households were screened and interviewed for the NSCH either on the Web, through the mail, or on the phone. In addition, within each modes respondents were randomly assigned an abbreviated or full length version of the survey. We compared eligibility, response rates, and health conditions reported by mode and considered the effects of questionnaire length in the models. The results shed light on sample variation, household size, reported health conditions, and eligibility all by mode and by questionnaire length providing researchers with considerations for the survey design.
Year of publication2016
Bibliographic typeConferences, workshops, tutorials, presentations

Web survey bibliography - The American Association for Public Opinion Research (AAPOR) 71st Annual Conference, 2016 (107)