This contribution reports from an academic rather than a research council viewpoint, and takes as its focus my experience in chairing the European Science Foundation review panel for the EUROCORES cross-national and multi-level Human Values, Institutions and Behaviour (HumVIB) programme. This programme was centred around the use of quantitative social science methods for exploring changes in views and attitudes, and my experience on the UK ESRC’s Research Grants Board as a statistician was thought to be relevant. The HumVIB research programme had specific and unique elements – proposals were sought for both standard research programmes of work but also two “research and support” packages, one developing and providing support for data construction and archiving, and the second providing methodological support and development in the statistical technique of multi-level models