Answers are grounded in this paper and cite their sources.
This study investigates whether it is the case that representativity is undermined if personal computer, tablet and smartphone respondents differ in socio-demographic characteristics and display different survey completion rates. Online market research is struggling with sample representativity. The analysis of more than ten million survey invitations, as well as stated device preference information, suggests that web survey respondents who are members of online panels still mostly use their personal computers, but do express increasing interest in using smartphones and tablets. Survey completion rates do vary across devices, and device use is significantly associated with socio-demographic characteristics and length of membership on a panel. Therefore, researchers must not limit respondents to use a specific device for completing a survey as this may compromise the quality of the survey completion experience, increase non-response error and negatively affect representativity.
No takes yet. Share an insight, caveat, or question.
Brosnan et al. (2017) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: