Figure: Percentage of all authors that are women who published in Perspectives (Science) and News & Views (Nature). The red lines indicate the proportion of women scientists within the field ^[1].

Great strides have been made in gender equality over the last decades. Today women have access to the same facilities and learned societies as men. In many countries the proportion of women to men earning undergraduate degrees in science and technology are approximately equal ^[1]. Programs are in place to address the disparity in the proportion of women in science and to ensure equality at all levels. However, there are still areas where gender bias occurs. 

It is well known that the proportion of women decreases drastically between the Ph.D. and the professor level in most fields. Most men perceive that the “leaky pipeline” is the current cause for gender inequality and it will solve itself with time as the proportion of women grows at the lower levels ^[2], although experience during the last decades does not support this view ^[3]. Women faculty earn less, are promoted less frequently to senior academic ranks, and publish less frequently than their male counterparts ^[4]. 

The occurrence of gender bias has not been considered previously in a systematic fashion in some areas. One of these areas is inclusion of women scientists at the highest levels, for example, as plenary speakers or in high-profile invitation-only workshops. We have examined two prominent, high-impact journals (Nature and Science) that have sections that are entirely, or to a large degree, contributions by invited authors only. The articles are published with the aim of placing recently published findings into a wider perspective. 

The data
We examined the gender distribution of authors of News & Views in Nature and Perspectives in Scienceduring 2010 and 2011. Volumes 463-480 (Issues 7277-7378) were examined from Nature and Volumes 327-334 (Issues 5961-6063) from Science. Contributions written by the editors were disregarded, as were other invitation-only sections, such as Retrospectives (Science), Obituary (Nature), and 50 & 100 years ago (Nature). The gender of the authors of 605 articles in Perspectives and 635 articles in News & Views was determined. In addition, we examined the gender distribution of authors from the Insight section inNature during 2009-2011 (Volumes 457-480). 

The gender of the authors was determined through internet searches on the affiliation addresses and research supjects. If no photos were found, or if the gender was not obvious to us, we contacted the authors and asked. The gender of two of the authors in both Perspectives and in News & Views are still unidentified. About one-half of the articles were written by full professors, but the authors were as often from a lower rank in the academic hierarchy (Ph.D. Student, Post-doc, Assistant Professor, or Associate Professor) or from outside of academia. 

We divided the articles into three broad areas: 1) Biological and Chemical Sciences which includes Medical Sciences; 2) Physical Sciences; and 3) Earth and Environmental Sciences. We merged Biological Sciences and Chemical Sciences articles into one group due to the relatively low numbers of chemistry articles and due to the difficulties in separating articles from the fields of biology, chemistry and medicine. The majority of articles (59%) in both Perspectives and News & Views sections were in the fields of Biological and Chemical Sciences. 

We found the lowest percentage of women authors in Earth and Environmental Sciences both from News & Views in Nature (3.8%) and Perspectives in Science(5.4%), with the greatest percentage of women authors (19%) in Biological and Chemical Sciences from Perspectives in Science (Figure). 

The proportion of women as authors in the Insightsection of Nature increased from 8% during 2004-2005 (134 authors) ^[5] to 18% during 2009-2011 (223 authors). There were too few articles in the Insight section to obtain reliable estimates for the different supject categories. 

Not all fields of science are equivalent in current gender representation. Therefore, we compared the proportion of women authors in News & Views andPerspectives with the proportion of women scientists employed in the U.S. in Science & Engineering occupations ^[1] (see Figure). The U.S. figures were used as a first approximation of the number of women employed in science jobs. The proportion of women as authors was greatly underrepresented when compared to the proportion of women scientists. The proportion of women authors in Biological and Chemical Sciences and Physical Sciences were about half compared to the proportion of women scientists, whereas the proportion of women authors in Earth and Environmental Sciences was up to 4- to 5-fold smaller than the proportion of women scientists. 

The consequences
The consequences are that women are not as visible as men and are not provided the same opportunities for career advancement, which contribute to the loss of many women in traditional academic tracks or from science all together. This loss of women in sciences constitutes a brain drain for society and a lack of role models for younger women in science ^[2]. An invitation as a keynote speaker at a large conference or a publication in a high-impact journal, makes an excellent contribution to your CV. Therefore, the opportunity to write invitation-only articles is of great importance to one’s science career. 

Gender bias is a larger problem in science. While gender parity in keynote speakers is achieved at most large science conferences, the gender ratio of the invited speakers at smaller conferences can be highly skewed, with few women appearing as an invited speaker. Gender parity can be achieved if a policy is implemented that strives for gender parity. An excellent example of a willingness to change isNature, because it increased the proportion of women authors in the Insight section after being criticized for offering authorship to too few women ^[3]. 

Fixing the problem
Recently, Ceci and Williams ^[6] proposed there is no sex discrimination in hiring practices, reviewing of journal articles, or in decisions regarding funding. They argue that additional root causes contribute to the low proportion of women in science due to differences in resources, abilities, and lifestyle choices ^[6]. Culture-based norms and prejudices regarding the role of women in society can create pervasive intangible barriers that hinder the inclusion of women ^[7]. Others have recognized there are structural barriers within academia slowing the advancement of women into higher levels, including the lack of mentors and role models and the lack of leadership training ^[8]. Strategies and guidelines have been developed to reduce this loss of women from science ^[9]. 

Why does gender bias occur in invitation-only activities as observed in these special sections ofNature and Science? The possibility for women to publish in general is as high as the possibility for male scientists, and discrimination has not been observed in the peer review process ^[6]. In addition, many of the section editors for Nature and Science are women, so we can exclude the possibility that it is only men making the decisions. For the most part women are no longer actively excluded in science only because of their gender; therefore we posit that it is unintentional bias. Discrimination can be remedied by looking beyond the boundaries of the present networks of the editors. 

Examination of the proportion of men and women that are invited to participate in all areas of science, whether that is as an invited speaker, a workshop participant, or as an invited author for Nature andScience is only good scientific practice. We can increase the proportion of women at all levels of science if we make the effort. 

Daniel J. Conley is Professor and Johanna Stadmark is a Researcher at the GeoBiosphere Science Centre, Department of Geology, Lund University, Sweden.
e-mail: daniel [dot] conley [at] geol [dot] lu [dot] se (daniel[dot]conley[at]geol[dot]lu[dot]se) andjohanna [dot] stadmark [at] geol [dot] lu [dot] se (johanna[dot]stadmark[at]geol[dot]lu[dot]se) 

References

1. National Science Foundation Scientists and Engineers Statistical Data System (SESTAT). Available athttp://www.nsf.gov/statistics/wmpd/tables.cfm.
2. Holmes, M. A. et al. Nature Geosci. 1, 79-82 (2008).
3. Thompson, L. et al. Nature Geosci. 4, 211-212 (2011).
4. National Science Foundation. Gender Differences in the Careers of Academic Scientists and Engineers: A Literature Review. NSF 03-322, Arlington, VA (2003).
5. Conley, D. J. Nature 438, 1078 (2005).
6. Ceci, S. J. & Williams, W M. Proc. Natl. Acad. Sci. USA 108, 3157-3162 (2011).
7. Filipsson, H. F. Nature GeoSci. 4, 346 (2011).
8. Etzkowitz, H. et al. Athena Unbound: The Advancement of Women in Science and Technology, Cambridge University Press (2000).
9. PRAGUES (2009) Guidelines for Gender Equality Programmes in Science. (link to pdf version).