International Women in Science Conference
Gerlind Wallon and Anna Ledin
Gender Blinding and the Long-term Fellowship Study
EMBO has monitored its selection processes with regards to gender balance since 2001, following the publication of a position paper on women in science. On average female success rate for applications to the EMBO Long-term Fellowship Programme is 20% lower than that for men. This difference has persisted despite the EMBO Fellowship-committee's awareness and best efforts.
We have therefore embarked on a serious investigation of the reasons for this persistent difference in success rate. A study by Wennerås and Wold, published in Nature in 1997, showed that in the case of the former Swedish MRC a strong committee bias led to a significant disadvantage of female applicants. In order to be able to glean some information about the further careers of our applicants, we chose to study the cohort of 1998. We analysed the full bibliometric data of all 680 applicants up to the beginning of 2006.
We noticed that successful female fellows were, bibliometrically speaking, slightly stronger then their male counterparts, but female applicants overall were slightly weaker. Was this difference sufficient to explain the observed difference in success rate? In order to test the influence of committee bias in general, we gender-blinded the committee members for the two rounds of application in 2006. Surprisingly, the difference in success rate persisted and even increased. We therefore concluded that the committee does not introduce a gender-based bias into the selection and that it must be aspects of the application itself that lead to the difference in outcome for men and women. We further investigated the difference in bibliometric quality of the applicants in 1998 and checked if this could in part be due to childcare responsibilities, the fact that more women applied from scientifically weaker countries or perhaps to an age/experience difference. Taken together these points can, to a large extent, explain the difference in bibliometric quality. We tried to define why not more of the highly qualified women were chosen, but investigation of the data for applicants around the cut-off value for selection remained inconclusive.
In summary, it can be concluded that the EMBO committee does not introduce a gender bias into the selection system, but that apparently the outcome of the selection procedure is influenced by the fact that female applicants accumulate a number of disadvantages. These disadvantages are apparent in the bibliometric quality of the female applicants which is on average slightly lower then that of the male applicants. Comparing the bibliometric data from 2006, it was very apparent that women had lost more ground. We followed this up by asking applicants to complete a questionnaire in order to find out why. In summary, women more frequently have a partner with a PhD degree, they are more likely to follow their partners, women work fewer hours then their partners and they earn the lesser part of the total family income. Children have a large negative effect on productivity for women, but surprisingly the opposite for men.
|A persistent problem - Traditional gender roles hold back female scientists
Anna Ledin, Lutz Bornmann, Frank Gannon & Gerlind Wallon