Essays

Data-driven predictions in the science of science

+ See all authors and affiliations

Science  03 Feb 2017:
Vol. 355, Issue 6324, pp. 477-480
DOI: 10.1126/science.aal4217

Figures

  • Fig. 1 How unexpected is a discovery?

    Scientific discoveries vary in how unexpected they were relative to existing knowledge. To illustrate this perspective, 17 examples of major scientific discoveries are arranged from the unanticipated (like antibiotics, programmable gene editing, and cosmic microwave background radiation) to expected discoveries (like the observation of gravitational waves, the structure of DNA, or the decoding of the human genome).

    GRAPHIC: ADAPTED BY K. SUTLIFF/SCIENCE
  • Productivity peaks early for most researchers.

    (Left) A heatmap showing the timing of the most productive year (measured in number of published papers) in a faculty career for more than 2300 computer science faculty, arranged from left to right by years since first faculty position (13). (Right) The histogram sums the heatmap’s rows, showing that, for most researchers, their most productive year occurred within 8 years of starting their laboratory.

    GRAPHICS: ADAPTED BY G. GRULLON/SCIENCE
  • Fig. 3 Major discoveries occur at any point in the sequence of a scientist’s publications.

    A raster plot showing the order of all publications, arranged from first publication to last, of 150 randomly chosen physicists (17), where each row of circles represents the sequence of publications by a particular scientist. Within a row, a blue circle marks the highest-impact publication. The uniform distribution of blue circles across the square, and the flatness of the corresponding histogram for 10,000 investigators (top), indicate that there is no pattern across the sequence as to when a major discovery occurs.

    GRAPHICS: ADAPTED BY G. GRULLON/SCIENCE
  • Who will publish the next breakthrough? Who will get grants? Who will get tenure?PHOTO: SEBASTIEN BONAIME/ALAMY STOCK PHOTO