Page - (000264) - in Knowledge and Networks

260 shows similar distributions for median conventionality (Fig. 12.11c) and papers with team authors show greater novelty than solo-authored or pair-authored papers (Fig. 12.11d). We plot the average hit citations by considering the top 5 % of highly cited papers written in the fields of geography, economics, and physics over time (Fig. 12.12). The average of hit citations for geography is significantly lower than that of eco- nomics and physics. For papers published between 1980 and 2000, half the papers in geography have median z-scores above 36. Thus papers in geography combine less conventionality when compared to physics or economics, where half the papers have median z-scores above 145. Moreover, for geography, conventionality doesn’t increase with time when compared with physics and economics (Fig. 12.13). This indicates that for geography, mixing novel and conventional combinations does not result in high impact work. Solo Author, High Tail Novelty Pair Authors, High Tail Novelty Pair Authors, Low Tail NoveltySolo Author, Low Tail Novelty 0.120 0.140 0.100 0.080 0.040 0.060 0.020 0.000 5 15 6545 7555 85 999525 35 5 15 6545 7555 85 99 BAH 0.120 0.140 0.100 0.080 0.040 0.060 0.020 0.000 Percentiles of Median z-score Percentiles of Median z-score 9525 35 Team Authors, High Tail Novelty Team Authors, Low Tail Novelty C Percentiles of Median z-score 5 15 6545 7555 85 999525 35 Fig. 12.9 Novel and conventional combinations in the production of science. The interplay between tail novelty, median conventionality, and hit paper probabilities show remarkable empiri- cal regularities (a–c). First, high tail novelty papers have higher impact than low tail novelty papers at (i) any level of conventionality and (ii) regardless of authorship structure. Second, increasing median conventionality is associated with higher impact up to the 85–95th percentile of median conventionality after which the relationship reverses. Third, larger teams obtain higher impact given the right mix of tail novelty and median conventionality. Nonetheless, at low levels of median convention and tail novelty, even teams have low impact, further emphasizing the fundamental relationship between novelty, conventionality, and impact in science. From Uzzi et al. (2013a, p. 471). Copyright 2013 by Science. Reprinted with permission from the authors and Science S. Mukherjee et al.