liu.seSearch for publications in DiVA
Change search
ReferencesLink to record
Permanent link

Direct link
On the cost-efficiency of probability sampling based mail surveys with a web response option
Linköping University, Department of Mathematics, Statistics. Linköping University, The Institute of Technology.
2005 (English)Doctoral thesis, comprehensive summary (Other academic)
Abstract [en]

Web surveys - the collection of survey data by means of self-administered questionnaires on the Internet - are becoming increasingly popular, perhaps mostly because of advantageous cost factors. If e-mail contacts are used, the cost benefit of a Web survey, compared with a mail or telephone survey, could be major. The reverse of the medal is that Web access is far from universal, and hence Web surveys only permit inference to populations of Web users. However, the Web can be used in combination with traditional response modes in a mixed-mode approach. Such an approach enables data collection from a probability sample of a general population, while still aiming at reducing costs by using the Web.

This thesis focuses on mixed-mode approaches in which (1) mail contacts are used and (2) a choice between a mail questionnaire and a Web questionnaire is offered. In the first paper, a theoretical framework is developed, enabling comparisons between different mixed-mode approaches with respect to minimum expected cost, subject to a precision requirement. A limited numerical study is performed to examine how the cost is influenced by the Web access rate and the Web response propensity. In the second paper, three mixed-mode approaches are evaluated empirically with respect to cost-efficiency. The mixed-mode approaches are compared with a mail-based reference approach with no Web response option. The cost-efficiency analysis is based on the results of a survey experiment, conducted in a population of university students in spring 2004. It is concluded that each of the three mixed-mode approaches was more cost-efficient than the reference approach.

Place, publisher, year, edition, pages
Linköping: Linköpings universitet , 2005. , 11 p.
Linköping Studies in Statistics, ISSN 1651-1700 ; 6
National Category
URN: urn:nbn:se:liu:diva-28448Local ID: 13595ISBN: 91-85299-39-1OAI: diva2:249257
Public defence
2005-05-26, Sal Planck, Fysikhuset, Campus Valla, Linköping, 10:15 (Swedish)

Papers included in this doctoral thesis are not included.

Available from: 2009-10-09 Created: 2009-10-09 Last updated: 2012-12-13

Open Access in DiVA

No full text

Search in DiVA

By author/editor
Werner, Peter
By organisation
StatisticsThe Institute of Technology

Search outside of DiVA

GoogleGoogle Scholar
The number of downloads is the sum of all downloads of full texts. It may include eg previous versions that are now no longer available

Total: 138 hits
ReferencesLink to record
Permanent link

Direct link