The objectives of the project are:
- To explore the opportunities that exist for matching auxiliary data from three different sources – small-area administrative data, commercial marketing data, and geocoded information from the Ordnance Survey on the physical location of sampled addresses – to data from the European Social Survey (ESS) in the UK. Each data source will be evaluated in terms of:
- What information is available which may be linked to the survey records of respondents and non-respondents;
- Whether accurate data is available for all sampled households in the UK;
- How successfully the risks of identification and disclosure which arise from combining data from multiple sources can be minimised whilst retaining information for analysis
- To investigate the information that auxiliary data can provide about potential biases present in the survey data as a result of nonresponse and how these biases might vary geographically across the UK. This will involve consideration of two distinct but related questions:
- To what extent can auxiliary data be used to predict sampled households’ propensity to respond?
- How are auxiliary variables shown to predict nonresponse related to the attitudes and behaviours the survey is intended to measure?
- To assess whether weighting using auxiliary variables to adjust for the possible over or under representation of certain types of respondent in the final survey:
- has a significant effect on survey estimates for different variables
- leads to a reduction in nonresponse bias