Source-linked AI summary
The Chaperone Effect in Scientific Publishing
Vedran Sekara, Pierre Deville, Sebastian Ahnert, Albert-László Barabási, Roberta Sinatra, Sune Lehmann
TL;DR
Quantitative evidence on how journal-specific publishing experience relates to scientists’ transition to senior authorship has been limited. This paper quantifies that “chaperone effect” across fields and finds it strongest in biology and general-topic journals, with chaperoned authors’ papers having greater impact than papers by new PIs.
Problem
Quantitative understanding of how apprenticeship relates to scientific publishing and the transmission of scientific excellence between generations is limited.
Source: p. 1Method
The paper quantifies the chaperone effect by comparing the observed ratio of new to chaperoned authors with a ratio from randomly permuted author orders.
Source: p. 3Results
The effect is strongest in biology and general-topic journals, and chaperoned authors’ papers have greater impact than papers by new PIs.
Source: pp. 3–4Takeaways & Limitations
The findings indicate that junior scientists learn the ability to publish in certain venues from senior colleagues.
Source: p. 4The analysis averages across authors with different levels of success and does not account for good protégés finding good mentors or young authors leaving science.
Source: p. 4
Abstract
from arXiv · showhide
Experience plays a critical role in crafting high impact scientific work. This is particularly evident in top multidisciplinary journals, where a scientist is unlikely to appear as senior author if they have not previously published within the same journal. Here, we develop a quantitative understanding of author order by quantifying this 'Chaperone Effect', capturing how scientists transition into senior status within a particular publication venue. We illustrate that the chaperone effect has different magnitude for journals in different branches of science, being more pronounced in medical and biological sciences and weaker in natural sciences. Finally, we show that in the case of high-impact venues, the chaperone effect has significant implications, specifically resulting in a higher average impact relative to papers authored by new PIs. Our findings shed light on the role played by experience in publishing within specific scientific journals, on the paths towards acquiring the necessary experience and expertise, and on the skills required to publish in prestigious venues.
Data.
The study uses Web of Science records and Nature’s publication history to analyze publications from 1960–2012 in interdisciplinary journals and five scientific fields. After filtering publications and journals, the final data cover six journal groups.
Data.
- Web of Science records include publication dates, journal and issue, author names and order, affiliations, and indexed references; Nature’s full publication history was also downloaded via its opensearch API.The database covers scientific outputs from 1898–2012 across more than 22,000 journals and contains more than 50 million papers.Source: Data · p. 4
- The analysis covers Nature, Science, and PNAS, plus journals in medicine, biology, mathematics, chemistry, and physics, identified by matching field lists to database journals.The study identified 97 biology, 337 medicine, 243 physics, 248 mathematics, 138 chemistry, and 3 interdisciplinary journals before further filtering.Source: Data · pp. 4–5
- The study analyzed categorized journals with publications spanning at least 10 years and at least 1,000 collected publications overall, retaining Articles, Letters, and Reviews while excluding titles with specified terms.The excluded title terms were “comment,” “reply,” “errata,” and “retracted article.”Source: Data · p. 5
- After preprocessing, the dataset comprised 795,558 biology, 1,350,936 medicine, 1,753,641 physics, 208,223 mathematics, 1,341,150 chemistry, and 251,294 Nature, Science, and PNAS publications.The first five fields drew on 40, 128, 117, 26, and 72 journals, respectively; the interdisciplinary publications came from Nature, Science, and PNAS.Source: Data · p. 5
Author name disambiguation.
Author names were standardized as lowercase surname-and-initial strings, and same-journal publication sequences with identical strings were treated as belonging to one individual.
Author name disambiguation.
- Names were lowercased and reduced to surname plus the first letter of each given name.For example, “John Smith” became “smith,j” and “Mary Suzy Johnson” became “johnson,ms”.Source: Author name disambiguation · p. 5
- Publications in the same journal with identical formatted author names were treated as authored by the same individual.Source: Author name disambiguation · p. 5
- Homonym errors were expected to be rare because they require distinct people with identical formatted names to publish in the same journal, an accurate field-level disambiguating feature (33).Source: Author name disambiguation · p. 5
Robustness of results to alphabetic ordering.
The study’s conclusions are robust whether alphabetically ordered publications are included or excluded, and observed C differs significantly from the alphabetical null model C_alphabet.
Robustness of results to alphabetic ordering.
- Across both the full and alphabetically filtered datasets, observed C differs significantly from C_alphabet, and the conclusions remain robust.Excluding alphabetically ordered publications removes 17.7% of Biology, 14.4% of Medicine, 30.9% of Physics, 75.1% of Mathematics, 23.3% of Chemistry, and 20.8% of interdisciplinary publications; these counts include both chosen and chance alphabetical ordering.Source: Robustness of results to alphabetic… · p. 5