Web Survey Bibliography
Under the project of 'Joint Research Activities (JRA2): improving representativeness of the samples’, one of the goal of this project is to study non-response bias using four
different approaches for detection and adjustment of nonresponse bias used in previous rounds of ESSi: (1) post-stratification; (2) analysis of reluctant respondents; (3). using
observable data; and (4) estimating bias and adjustment using surveys among nonrespondents. These four approaches on detection and adjustment of non-response bias
are used independently from each other and each approach has its own advantages and disadvantages. The last approach applied propensity score stratification method (10
classes) on the basis of the information of co-operative respondents and non-respondents. The weighting, based on this method, was not very effective. Following the adjustment
methods in the last approach, this paper reflects and explores other adjustment methods rather than using propensity score stratification or direct estimation methods. Propensity score methods based on matching, inverse probability of treatment weighting (IPTW) or 'multiple treatment approach’ including information of reluctant respondents in addition to cooperative respondents and non-respondents are possibly considered in this paper. The paper also wishes to contribute discussion on the integration of four aforementioned approaches studied under JRA2.
Conference Homepage (abstract)