Statistical Physics Group

A new agent-based model which explains the relationship between research-group quality and quantity is described. Statistical analyses of RAE results allow measurement, for the first time, of critical masses in research. For the mathematical sciences, these are rather low in comparison with other disciplines, the smallest critical masses being for pure mathematics. A snapshot of the relative health of research-group structure in pure and applied mathematics as well as in statistics and operational research in the UK is given. A cautionary note against the usage of onesize- fits-all minimum group sizes for quality-related funding is issued as this could have a devastating effect in the mathematical sciences. A caution is also issued regarding the absence of normalisation of peer-review measures of research quality across disciplines as this has significant consequences for fair allocations of funding.