Most articles today are results of teamwork, whether it’s only two authors working together or thousands, (think CERN). As science keeps getting bigger, authorship no longer equals actual writing, but one way or another of contribution to team effort. Authorship of massive scale, or “Hyperauthorship” (Cronin, 2001) is very common in high-energy physics and certain sub-fields of biomedicine.
The impact of team authorship
A 2007 Science article (Wuchty et al.) analyzed 19.9 million research articles taken from the Web of Science (WoS) database, as well as in 2.1 million patent records. Highly cited work was considered as one which received “more than the mean number of citations for a given field and year“. I have a bit of a problem with this definition, since it gives the same weight to a work which received 8 citations in a field where the mean is 3 and to a work in the same field which received 50.
They defined a measurement of relative team impact (RTI) for different periods and fields. RTI is “the mean number of citations received by team-authored work divided by the mean number of citations received by solo-authored work.” The RTI was bigger than one for all periods and all large research areas, and grew over time. In science and engineering, citations for team articles increased from an average of 1.7 times those of single-authored articles in 1955 to 2.1 times in 2000.
To make sure the teams’ advantage didn’t come from self-citations, they ran the analysis again without them. Removing self-citations reduced the RTI by 5%-10% (depends on the field), but the relative advantage of teams remained more or less the same. Not only that, but team-authored articles also have a bigger chance of being top-cited than single-authored articles. A team-authored article in science and engineering is 6.3 times more likely to receive more than 1,000 citations.
Aksnes (2003) looked at highly cited papers from the National Science Indicators (NSI) and the National Science Report (NCR) for Norway between 1981 and 1996. His definition for a “highly cited paper” was one that was cited “at least 17 times the mean NSI citation rate in the particular subfield/year.” The average number of authors per article in Aksnes’ sample was 8.9, but he mentions that the results are skewed due to 4 articles with more than 200(!) authors. The average for the rest of the publication set was 3.7 authors per article. It’s worth noting that, unlike Wuchty et al., Aksnes chose to remove the art and humanities category from his analysis because bibliometrical databases mostly cover journals, and publishing in those disciplines is usually done through books and monographs. The increase in authorship Wuchty et al. showed in art and humanities (not too big anyway) could have been, in theory, cancelled out if they were able to include those kinds of publications.
Increased number of authors can mean dilution of credit. In a survey of medical schools’ promotion committees, a majority of respondents agreed that in a three-author byline, the first did most of the work, while the last deserved most of the credit for the idea and supervision. The last author’s perceived credit didn’t change in a five-author byline, but the first author’s part of the idea credit decreased from 37% to 29% (Wren et al., 2007). I don’t want to know what happens in an article of 300 authors. Another study (Costas and Bordons, 2011) found what everyone already knows: junior researchers tend to be first authors; senior, older researchers tend to be last.
The up side of co-authoring with a senior, famous scientist is that there’s a better chance somebody’s actually going to read the article. The down side is that your name will pale in comparison, because people give more credit to researchers who are already well-known. That is what renowned sociologist Robert K. Merton (1968) called “The Matthew Effect.” The name comes from the Gospel of St. Matthew: “For unto every one that hath shall be given, and he shall have abundance: but from him that hath not shall be taken away even that which he hath.” Essentially, it’s the academic version of “The rich get richer.”
Team work is the present and the future of science. However, it might be exploited for less-than-ethical purposes. In part II of this post we’ll talk more about the size of teams, why has team work become so popular, and the ethics of authorship.
Costas, R., & Bordons, M. (2011). Do age and professional rank influence the order of authorship in scientific publications? Some evidence a micro-level perspective. Scientometrics, 88, 145-161
Cronin, B. (2001). Hyperauthorship: A postmodern perversion or evidence of a
structural shift in scholarly communication practices?
JASIST, 52 (7), 558-569 DOI: 10.1002/asi.1097
Merton, R.K. (1968). The Matthew effect in science Science (159), 56-63
Wren, J.D., & et al. (2007). The write position. A survey of perceived contributions to papers based on byline position and number of authors. EMBO Rep, 8 (11), 988-991 DOI: 10.1038/sj.embor.7401095
Solomon J (2009). Programmers, professors, and parasites: credit and co-authorship in computer science. Science and engineering ethics, 15 (4), 467-89 PMID: 19247811
Wuchty S, Jones BF, & Uzzi B (2007). The increasing dominance of teams in production of knowledge. Science (New York, N.Y.), 316 (5827), 1036-9 PMID: 17431139