Good quality, good business
Improving Panel Sampling
Embedding Propensity Scores & Response Behavior in Sampling Frames
Lennart Huizing, Martin Müller, Clasine van der Wal, Marianne Hubregtse, Gerty Lensvelt-Mulders & Robert van Ossenbruggen
Abstract
- Sampling methods should take individual response inclination into account, because this varies between panel-members.
- Response inclination sampling yields intrinsically different people compared to propensity and quota sampling.
- Response inclination sampling improves efficiency compared to propensity sampling.
- Response inclination sampling improves validity compared to quota sampling.
- Where both validity and efficiency are wanted, response inclination is the best compromise.
< BACK
Lennart Huizing
Motivaction International, Netherlands
Graduated in Political Sciences at the Amsterdam University. Lennart currently manages research and development at Motivaction International. This department focuses among other things on panel and community management and optimisation and innovation of research processes. He is also a member of the management team. His interests lie in improving response rates and sampling methods for online research.
Martin MUller
Motivaction International, Netherlands
Martin graduated in Psychology at the University of Amsterdam, with specialization in Child- and Neuro-psychology. He worked at a medical research institute as statistician for 13 years and is Motivaction's senior methodologist since 2005. He is mainly interested in panel member behaviour, focussed on explaining panel membership, but also response and non-response issues of panel members.
Clasine van der Wal
Motivaction International, Netherlands
Graduated in Social Sciences in 2002 at the VU University Amsterdam. During her study she developed an interest in methodology. At the moment she works as a methodologist at Motivaction International, a research and strategy company based in the Netherlands. Her main interest lies in maintaining Internet panels and the data-quality from Internet samples.
Marianne Hubregtse
Utrecht University, Netherlands
Though originally a psychology student, the interest in methodology took charge. Now in the second year of a research Master on methodology and statistics at the Utrecht University, Marianne participated in this research for her internship. Her paper together with G.J.L.M. Lensvelt and P. Lugtig on non-response bias in panels has been submitted to Methodology. Plans for the future lie in the academic world, with interests in survey research with children under 18 years.
Gerty Lensvelt-Mulders
Utrecht University, Netherlands
Gerty is an associate professor at the department of Methodology & Statistics, Utrecht University, the Netherlands. Her main research interest is the improvement of the quality of access-panel surveys, by studying the relation between reliability and validity of non-random sampling designs. She works with government, business, and academic clients and regularly teaches short courses outside the university. As a hobby she works with young game-developers designing computer-assisted surveys for children in game-like environments.
Robert van Ossenbruggen
ProCression, Netherlands
Robert is an independent research consultant and an active member of the Dutch Marketing Research Association (MOA) as chairman of the Methodology Program Group and author for MOA's industry magazine Clou. He has been one of the initiators of the Dutch NOPVO study, an industry study among 19 online access panels in the Netherlands. Robert is highly interested in innovation in methods, techniques and applications.
Register to attend Panel Research 2007