Andrew Plume is Associate Director – Scientometrics & Market Analysis in Research & Academic Relations at Elsevier. Looking at a broad spectrum of data, he studies information flows in the scholarly literature by analyzing patterns of publications and citations. His particular interest lies in the use (and abuse) of the Impact Factor and the emergence of alternative metrics for journal evaluation.
Why scientists don't share and why they should
By Andrew Plume, Elsevier
Historically, studies have shown and continue to demonstrate that researchers desire to disseminate information and further knowledge within their disciplines. But they are also fundamentally human and susceptible to the drivers that motivate us all, including advancement and competition. In a 2009 article, Mulligan and Mabe report the key motivation for publishing — to disseminate results — supports the popular wisdom that academic research is a culture of sharing.1 But as the secondary (and, the study authors argue, more “covert”) motivations show, there are also deep-seated reasons why scientists do not share, including future funding, recognition and establishing precedence.
Decreasing levels of funding
Decreasing levels of institutional funding are a reality in many countries around the world, and the effect may be disproportional for early career researchers. From 2000 to 2008, success rates dropped (and not just due to increasing numbers of applicants) for the National Institutes of Health R01eqivalent grants, while at the same time the average age of awardees increased (Figure 1). This trend exacerbates pressure on researchers who, particularly in the early stages of their careers, are driven by the need to further themselves.2
Protection and prestige
Reasons for withholding research data may extend beyond personal considerations. Researchers may feel bound to protect the interests of others, such as a graduate student or post-doc, and sharing may also be prohibited by confidentiality agreements with government or industrial sponsors. Survey results show that 74% of scientists, on average, believe that access to other researchers’ data benefits — or would benefit — their own research; however, only 54% are willing to allow other researchers access to their raw research data.3 And the gap is increasing over time. Academics may also be disinclined to share and collaborate because of the nature of the reward structure in science, which esteems first discoverers.
A fine balance
Collaboration and competition go hand-in-hand and are driven mainly by the same forces, including funding; relationships (allegiances and animosities); ideological, social and political factors; and the publish or perish culture. Modern collaboration has evolved into a fine balance, trading off the inevitable sharing of credit against the benefits that accrue to each individual researcher from participating in a larger collaborative group. Beyond altruistic goals for society, collaboration can also accrue personal rewards for the scientist. Multi-authorship has been shown to increase citation rates, and international collaboration is responsible for a further boost in citation impact (see Llewellyn Smith story) over and above a simple multi-authorship effect.
Researchers’ endeavors begin by studying and building off the work of others. A natural progression in knowledge acquisition and creation leads those same individuals to eventually connect with other researchers — connections that can lead to new synergies in adding to a discipline’s body of knowledge, and open doors to new sources of funding and career opportunities.
(1) Mulligan, A., and Mabe, M.A. (2011), “The effect of the Internet on researcher motivations, behaviour and attitudes,” Journal of Documentation, Vol. 67 No. 2, pp. 290-311.
(2) Laudel, Grit, and Gläser, Jochen (2008), “From apprentice to colleague: The metamorphosis of Early Career Researchers,” Higher Education, pp 387-406.
(3) Campbell, E.G., Clarridge, B.R., Gokhale, M., Birenbaum, L., Hilgartner, S., Holtzman, N.A., Blumenthal, D. (2002), “Data Withholding in Academic Genetics, Evidence From a National Survey,” Journal of the American Medical Association, Vol. 287 No. 4, pp. 473–480.