Predicting Publication Success for Biologists

Roundtable Predicting Publication Success for Biologists William F. Laurance, D. Carolina Useche, Susan G. Laurance, and Corey J. A. Bradshaw Keywords: academic performance, gender, language, publication success, university ranking T he academic world can seem obsessed with metrics of a researcher’s performance, of which publication frequency is among the most important (Fischer et al. 2012). Further measures of research impact, such as personal citation rates and journal rankings, are also key indicators of academic success (Symonds 2004, Hirsch 2005, 2007, Acuna et al. 2012). For researchers, two realities seem unlikely to change in the foreseeable future. First, individual scientists vary greatly in their publication rate (Allison and Stewart 1974). Second, employment opportunities, grant success, and professional accolades are often tied intimately to one’s publication prowess (Zuckerman 1967, Reskin 1977, but see Leahey 2007). Given these realities, we sought to determine whether one can predict who is likely to publish prolifically during his or her scientific career and who is not. We tested five characteristics that might plausibly be used to gain insight into an academic’s future publication success. The first factor that we assessed was gender. We included this factor because prior studies have suggested that female scientists generally produce fewer publications than do male scientists and are more poorly represented on the upper rungs of the academic ladder (Long 1992, Holt and Webb 2007, Ceci and Williams 2011, Dugdale et al. 2011). Many possible explanations have been considered for this trend, including the heavy demands of motherhood during the crucial early phases of a woman’s career (Monosson 2008, McGuire et al. 2012, O’Brien and Hapgood 2012), potential gender bias (Brown 2008), a tendency for women to avoid self-promotion (Moss-Racusin and Rudman 2010), and feelings of isolation among female researchers (Nature 2011), among other reasons (see also Laurance et al. 2011). The second factor was whether English was the researcher’s native language. English has become the dominant language of international science; for example, 87% of all papers listed in Biological Abstracts are in English, with no other language constituting over 2% (Monge-Nájera and Nielsen 2005). For those for whom English is not their first language, proficiency varies greatly and often influences their attitude toward publishing in English-language journals (Ferguson et al. 2011). To be competitive for high-ranked journals, those with limited English skills may be forced to collaborate with native English speakers, to use commercial editing services, or to have their key works translated into English (Meneghini and Packer 2007, Primack and Marrs 2008). A third predictor was the overall prestige of the university from which the researcher received his or her doctorate. This factor could show effects if higher-ranked universities attract better, more-motivated students, who are more likely to succeed in the long term. Potentially contributing to this are greater financial resources for research and a “culture of success” evident at prestigious universities. In most metrics in which universities are compared, the number BioScience 63: 817–823. ISSN 0006-3568, electronic ISSN 1525-3244. © 2013 by American Institute of Biological Sciences. All rights reserved. Request permission to photocopy or reproduce article content at the University of California Press’s Rights and Permissions Web site at www.ucpressjournals.com/ reprintinfo.asp. doi:10.1525/bio.2013.63.10.9 www.biosciencemag.org October 2013 / Vol. 63 No. 10 • BioScience 817 Can one foresee whether young scientists will publish successfully during their careers? For academic biologists on four continents, we evaluated the effects of gender, native language, prestige of the institution at which they received their PhD, the date of their first publication (relative to the year of PhD completion), and their pre-PhD publication record as potential indicators of long-term publication success (10 years post-PhD). Pre-PhD publication success was the strongest correlate of long-term success. Gender, language, and the date of first publication had ancillary roles, with native English speakers, males, and those who published earlier in their career having minor advantages. Once these aspects were accounted for, university prestige had almost no discernable effect. We suggest that early publication success is vital for aspiring young scientists and that one of the easiest ways to identify rising stars is simply to find those who have published early and often. Roundtable A survey of biological and environmental scientists At the outset, we surveyed 1447 academics at 35 universities in North America (Canada, Mexico, United States), South America (Brazil, Chile, Colombia, Ecuador), Europe (France, Germany, Norway, Portugal, United Kingdom), and Australia. The academics were all in the biological and environmental sciences (i.e., zoology, botany, ecology, evolution, genetics, environmental science and policy). We excluded other fields of science, because they can have different publication rates and patterns of collaboration (Larsen and von Ins 2010), which might distort our findings. Of the 1447 potential candidates, we identified 182 who met two key criteria for detailed analysis. They (1) had completed their PhD before 2000 (giving us a 10-year window after the PhD to assess publication success) and (2) had an updated copy of their curriculum vitae (CV) available online (i.e., with information on their publication record, as well as data on gender, the year of PhD completion, and the university from which the PhD was granted). For these candidates, we also determined (through their CV, e-mail correspondence, or our personal knowledge) whether English was their first language. We included candidates regardless of their tenure status. Notably, less than one-third (32.4%) of the researchers in the academic departments that we surveyed were female (range: 11%–62%), and so we attempted to reduce the numerical bias toward male subjects. To do this, we e-mailed 62 female academics whose PhD was granted before 2000 but whose CV was not provided online and requested an updated copy (prior studies have shown that female academics tend to have a personal homepage, which is often a source for an online CV, less frequently than do 818 BioScience • October 2013 / Vol. 63 No. 10 male academics; Barjak 2006). Of those who responded, 21 fit our selection criteria and were included in our sample of 182 subjects. The universities that we surveyed varied widely in their ratings of prestige (with rankings ranging from 1 for Harvard University to higher than 1200 for some universities). The rankings were based on the Academic Ranking of World Universities (ARWU; www.arwu.org). This index has been lauded for being consistent and transparent and f (...truncated)