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

Web Survey Bibliography

Title Comparing Propensity Score Weighting with Other Weighting Methods: A Case Study
Year 2003
Access date 07.05.2004
Presentation pdf (32k)
Abstract This paper deals with weighting issues in Web surveys of the general population. A common way of conducting such surveys is to recruit a volunteer panel of Web users from which quota samples are selected. Inference from such non-probability samples relies heavily on the existence of good weighting procedures. A weighting procedure examined here is the propensity score method, which with certain assumptions has a potential for reducing selection bias in survey estimates substantially. In this procedure, a parallel probability-based survey – usually a telephone survey – is used to estimate the propensities of being in the Web sample based on a vector of covariates (“lifestyle” variables) measured in both modes. Critical for the method is an appropriate choice of the life-style questions.
We compare the propensity score method with conventional weighting techniques for a data set collected in 2001 by the Danish marketing research company Zapera in four Nordic countries: Denmark, Norway, Sweden, and Finland. The data were in each country collected parallel from Zapera’s volunteer Web panels and from independent phone number samples. Two groups of variables were studied: (i) the use of various hygiene articles and (ii) attitudes to local banks. In both samples a number of “modernity” statements were added, which the respondents could agree or disagree to. These statements are used as lifestyle variables for the propensity score weighting.
Access/Direct link Homepage - conference (abstract)
Year of publication2003
Bibliographic typeConferences, workshops, tutorials, presentations
Print

Web survey bibliography - Other (439)

Page:
Page: