Working with mixed-mode survey data? Join this free webinar, run by CLS and Survey Futures, to understand the challenges of using mixed-mode survey data and learn statistical methods to handle these in practice.
About the event
Surveys are increasingly moving to mixed-mode data collection – such as carrying out interviews via face-to-face, telephone, video and/or web modes.
In this webinar, we will give an overview of issues that arise when using data collected in mixed-mode surveys. This includes the bias introduced when participants respond differently to survey items depending on the survey mode used – termed “mode effects”.
We will conceptualise the bias from mode effects within a simple and intuitive empirical framework called Causal Directed Acyclic Graph (DAG). We will then describe statistical methods for handling mode effects, looking in particular at Quantitative Bias Analysis (QBA).
Why attend?
- Learn about mixed-mode designs and the reasons they can introduce bias in data analyses.
- Find out how you can apply DAGs to easily conceptualise the bias from mode effects.
- Learn statistical methods for handling mode effects, their assumptions and the situations where they may increase bias.
- Learn about QBA.
Who should attend?
Users or managers of mixed-mode survey data, including users of CLS cohort data.
Presenters:
Liam Wright
Lecturer in Applied Statistical Methodology
Richard Silverwood
Associate Professor and Chief Statistician
Georgia Tomova
Research Fellow (Statistics/Quantitative Social Science)
