Stanley and colleagues looked at five different measures of research activity in universities: the annual expenditure on R&D, the number of papers published each year, the number of patents awarded, whether R&D funds had been transferred from one university to another by a funder, and the size of schools and departments within universities. They compared results from 719 universities in the US with data from the UK and Canada over a 17 year period.

They found that the growth in research activity did not depend on the size of R&D expenditure (i.e. the size of the university), just as business growth does not depend on the size of the company. The research also suggested that in order to make fundamental contributions to science, university research programmes should last more than five years - two years longer than the average research grant.