Slopes (b) and their standard errors (SE) from linear models relating the logarithm of total scientific impact to the logarithm of grant support received from Individual Discovery grants from the Natural Sciences and Engineering Research Council of Canada (NSERC) in 2002–2006.

Impact was measured for the period of 2003–2006 as: a) numbers of publications, b) numbers of citations to those publications by 2012, c) the number of citations received by the most highly cited paper, and d) the number of very highly cited papers. Relationships were determined for three disciplines: Integrative Animal Biology (Bio), Inorganic and organic chemistry (Chem) and Ecology & Evolution (Ecol). In two cases, a break point (B.P.) was estimated visually from LOWESS plots, and the slope estimated using only the points higher than the B.P. We tested slopes for significant difference from 0 (pb = 0) and from 1.0 (pb = 1). 0.050.10.