Activity Number:
|
462
- SPEED: Survey Research Methods
|
Type:
|
Contributed
|
Date/Time:
|
Wednesday, August 2, 2017 : 8:30 AM to 10:20 AM
|
Sponsor:
|
Survey Research Methods Section
|
Abstract #323373
|
View Presentation
|
Title:
|
The Link Between Nonresponse Bias and the Length of the Field Period in a Mixed-Mode General Population Panel
|
Author(s):
|
Bella Struminskaya* and Tobias Gummer
|
Companies:
|
Utrecht University and GESIS - Leibniz Institute for the Social Sciences
|
Keywords:
|
length of field period ;
risk of nonresponse bias ;
mixed-mode panel ;
online panel
|
Abstract:
|
Survey researchers are often confronted with the question about the length of the field period. A longer fielding time can increase participation but it requires more effort from survey managers. In this paper we study how sample composition changes during the field period. Our main research question is whether a longer fielding time reduces the risk of nonresponse bias. We use data from the GESIS Panel, a probability-based mixed-mode panel of the general population in Germany (N=4938). We analyze online and offline samples, in which respondents were invited to bi-monthly surveys with a field period of 2 months. Using the information from the recruitment interview and the first self-administered survey we predict panelists' response propensities. We then analyze how nonresponse bias indicators vary over the field period using coefficients of variation and R-indicators. We find differences in participation propensities by mode: underrepresented propensity strata begin to increasingly participate online after about 2 weeks and offline after about 40 days. Hence, decreasing the length of the field period would increase the risk of bias in the online mode more than in the offline mode.
|
Authors who are presenting talks have a * after their name.