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

Web Survey Bibliography

Title Evaluating Three Approaches to Statistically Adjust for Mode Effects
Year 2016
Access date 19.04.2016
Abstract A major hazard in conducting multimode surveys is the potential for mode effects to compromise the response distributions recorded. We evaluate the strengths and weaknesses of three approaches for statistically adjusting for mode effects. Under a regression modeling approach, adjustments are computed by regressing survey responses on mode, demographics, and other relevant variables. Under a multiple imputation approach, mode effects are conceptualized as a missing-data problem. The standard multiple imputation techniques such as chained equations can be used to impute the responses in additional modes. We also propose a new imputation approach based on an econometric framework of implied utilities in logistic regression modeling. We evaluate all three approaches using data from the second wave of the Portraits of American Life Survey sponsored by Rice University's Kinder Institute for Urban Research. This survey featured online and CATI interviewing with a national sample of adults, with random assignment to either CATI-only or web with CATI follow-up for nonrespondents. We develop a workflow to determine which variables require a mode effects adjustment based on standard false discovery rate multiple hypothesis testing procedures. We detected a significant mode effect on four survey outcomes after controlling for demographics and risk of type I error. The mode effects adjustments were then applied to these variables. The effects on the standard errors and point estimates are examined and discussed along with the advantages and disadvantages of each adjustment approach. The multiple imputation approach produced estimates with better apparent accuracy, as evidenced by better internal consistency of the estimates and a moderate increase in the standard errors. Unlike the regression adjustment approach, which can only produce aggregated estimates for the whole study, the multiple imputation approach can be used for disaggregated analysis with mode-adjusted estimates as well.
Year of publication2014
Bibliographic typeJournal article