It is well known that even the best conducted surveys often generate significant amounts of error during their design and implementation, best described by the Total Survey Error (TSE) framework. It is also widely accepted that cross-national surveys have the potential to increase that error further. This is because they most often comprise multiple national surveys under some kind of coordinated framework, whilst also having additional sources of error that stem from the cross-cultural nature of that work. Recently great strides have been made in identifying the sources of error that can impact on social surveys and how these are magnified in a cross-national context. This doctorate presents a body of my own published work that has contributed to the field of cross-national research. It has provided tools and approaches that help in the identification and correction of three overarching aspects of non-sampling survey error: specification error; measurement error and non response error. In each of these areas my contribution to the field through work on the cross-national European Social Survey (ESS) will be demonstrated, drawing on peer reviewed journal articles, book chapters, a book and published working papers. This academic contribution has added to knowledge in the field and made a practical contribution by leading to tangible improvements in the methodology of the ESS and other cross-national social surveys.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:685570 |
| Date | January 2015 |
| Creators | Fitzgerald, R. |
| Publisher | City University London |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://openaccess.city.ac.uk/14487/ |
Page generated in 0.0017 seconds