Page - (000261) - in Knowledge and Networks

257 Role of Teams in Production of Knowledge Collaboration is often claimed to produce more novel combinations of ideas (Falk- Krzesinski et al., 2010; Fiore, 2008; Stokols et al., 2008; Uzzi & Spiro, 2005; Wuchty et al., 2007), but the extent to which teams incorporate novel combinations across the universe of fields is unknown. Team-authored papers were more likely to show atypical combinations than single or pair-authored papers. Figure 12.8a shows that the distribution of 10th percentile z-scores shifted significantly leftward as the number of authors increased (Kolmogorov-Smirnov [KS] tests indicate solo vs. pair p = 0.016, pair vs. team p = 0.001, team vs. solo p < 0.001). Papers written by one, two, three, or more authors showed high tail novelty in 36.1 %, 39.8 %, and 49.7 % of cases, respectively, indicating that papers with three or more authors showed an increased frequency of high tail novelty over the solo-author rate by 37.7 %. Teams were neither more nor less likely than single authors or pairs of authors to display high median conventionality. Figure 12.8b indicates no significant statistical difference in the median z-score distributions (KS tests indicate solo vs. pair p = 0.768, pair vs. team p = 0.417, team vs. solo p = 0.164). Teams thus achieve high tail novelty more often than solo authors, yet teams were not simply “more novel” but rather displayed a propensity to incorporate high tail novelty without giving up a central tendency for high conventionality. Regression Methods In our final analysis, we examined the interplay between citation, combination, and collaboration using regression methods (Fig. 12.9). Papers were binned into eleven equally sized categories of median conventionality. We used logistic regression to Table 12.2 Novelty, conventionality, and citation impact by field Rank 1st 2nd 3rd 4th High tail novelty and low median conventionality 20.3 % 44.5 % 28.7 % 6.5 % Low tail novelty and high median conventionality 9.7 % 26.7 % 50.6 % 13.0 % High tail novelty and high median conventionality 64.4 % 21.9 % 3.6 % 10.1 % Low tail novelty and low median conventionality 5.7 % 6.9 % 17.0 % 70.4 % Note. For each of 243 subfields indexed by the WOS in the 1990s, we ranked the categories of papers according to their probability of producing hit papers. Hit papers are defined as those in the upper 5 % of citations received in that subfield. We focused on all papers published across all sub- fields in the 1990s. This analysis revealed that high tail novelty and high median conventionality were the highest impact papers in 64.4 % of subfields and either first or second in 86.3 % of fields. By contrast, low tail novelty and low median conventionality rank lowest or second lowest in 87.4 % of fields. From Uzzi et al. (2013a, p. 22). Copyright 2013 by Science. Reprinted with per- mission by the authors and Science 12 How Atypical Combinations of Scientific Ideas Are Related to Impact:…