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

Web Survey Bibliography

Title Differential Sampling Based on Historical Individual-Level Data in Online Panels.
Year 2011
Access date 29.07.2011
Abstract

Non-response bias can significantly compromise the representativeness of a sample. In order to combat these problems, researchers tend to mandate marginal quotas to known population proportions. However, the problems associated with quota sampling have been extensively documented when marginal quotas do not fill up at the same time. Differential sampling, whether through propensity scores or model-based response rate adjustments, avoids the problems involved in quota sampling. Differential sampling oversamples subpopulations with lower response rates to ensure survey respondents are balanced according to population proportions. Adjusting for these subpopulations becomes even more important in the online space where individual response rates vary dramatically.

Practitioners who use online panels must develop solutions for coverage limitations, conditioning and respondent attrition, source variability, and non-response bias. We discuss how to more effectively correct non-response bias in online sample to create more accurate estimators from data collected online. The most common way to adjust for non-response bias involves modeling the response rate for each of the subpopulations and oversampling accordingly. Some real time sources use aggregate level demographic estimates to adjust the flow of respondents into their surveys. However, demographics do not account for a large percentage of the variation in response rates. Panel companies can instead use historical response rate data at the individual level to accurately adjust the flow of respondents and correct for non-response. We discuss a system of implementing individual-level response rate estimation and show the advantages when compared to model-based response rate estimates.

 

Access/Direct link

Conference Homepage (abstract)

 

Year of publication2011
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web survey bibliography - The American Association for Public Opinion Research (AAPOR) 66th Annual Conference, 2011 (26)