Methods for combining probability and non-probability samples have been around for two decades or more, yet their use is far from widespread, especially in the UK. There is now growing interest in their potential, given the increasing costs and challenges of conducting social surveys.
About the event
In this webinar, we will present an overview of the range of methods available for integrating or combining probability survey samples with data generated via non-probability methods. This will include a discussion of the different data structures for which they were designed and the evidence regarding their respective strengths and limitations.
Following the overview of methods/approaches, we will focus on the practicalities of using them in the “real-world”, covering considerations such as the purpose of combining the data and how this relates to the choice of method. We will then outline the key steps in implementation, including selection of covariates, data reduction, and assessment of the success – or otherwise – of the adjustments made.
After a brief summary, there will be opportunity for questions and open discussion.
Why attend?
- Learn about the methods available for combining probability and non-probability samples.
- Find out more about how these methods are used in practice and what they can and cannot achieve.
- Discover more about why these methods and under-utilised and how this might change in future, especially in the UK.
Who should attend?
The workshop is aimed at a wide audience including survey methodologists, researchers, data users and research commissioners. There will be some discussion of technicalities, but the main focus is on the viability and practicalities of combining data from probability and non-probability samples.
Event timings
12:00 – Welcome and introduction
12:10 – Review of methods and their strengths/limitations
12:30 – Practical considerations and implementation
12:50 – Summary and areas for future research
13:00 – Q&A / discussion