Abstract:
|
Multi-mode surveys have been in use for a long time as survey managers seek to use collection procedures that produce the best possible data within existing constraints of time and budget (de Leeuw, 2005). Especially, Web or CATI surveys (laptop, tablet, pad, PC) become more and more popular due to people's willingness to adopt advanced technologies in their daily lives. Consequently, such multi-mode designs lead to a confounding of selection effects and measurement effects (measurement errors) caused by mode differences. This research will investigate the mode effects by developing propensity models of R-Indicators. R-indicators are designed to measure the similarity between respondents and the original sample or survey population. They are used to measure representativeness of respondents and to identify which subgroups are over- or under-represented. In this research, three multi-mode surveys are analyzed using the same propensity model.
|
ASA Meetings Department
732 North Washington Street, Alexandria, VA 22314
(703) 684-1221 • meetings@amstat.org
Copyright © American Statistical Association.