Fig. 1: Procedure for defining prizewinning topics.
From: Scientific prizes and the extraordinary growth of scientific topics
a Our procedure has three steps. First, we collected a large sample of over 400 recognized prizes recorded in Wikipedia. Second, we linked prizes to prizewinners using the prize’s homepage and the prizewinner’s Wikipedia page. Third, we linked topics to prizes via the prizewinner’s Known For (KF) topics. KF topics are the subset of topics that a researcher is known for out of all the topics a researcher has worked on. We defined a researcher’s KF topics as those topics on which the researcher had published 10 papers or more, which was cross-validated with Wikipedia’s KF topics pages. Wikipedia’s KF topic’s page uses crowdsourcing to identify a researcher’s KF topics (Robustness Check, Supplementary Fig. S1). For any known-for topic associated with a prize, we used MAG to verify that the prizewinner had publications on the known-for topic that predated the year of the associated prize. Each prizewinning topic was then matched with five non-prizewinning topics that had statistically indistinguishable year-to-year growth from the prizewinning topic during the 10-year period prior to the prizewinning year as explained in detail in the main text. b Δt measures the difference between the log of the average growth of prizewinning topics minus the log of the average growth of matched topics after the prize year. The inset shows Δt as the relative percentage growth of the prizewinning topic compared to the matched topic group, hence the time prior to prize year, t = 0 is flat. Δt is measured and reported separately for 6 growth measures of a topic as described in the text. Gray panel represents the time before the prize year. Illustration images are from www.pxfuel.com and www.pikpng.com.
