Gendered patterns in international research collaborations in academia

Abstract

Although women's representation in higher education nears parity with men at the undergraduate level, this representation diminishes as one ascends the academic ranks. Because gender gaps in the ‘elite’ activity of international research collaborations might contribute to the underrepresentation of women in the upper ranks, we ask if gender differences exist in participation in international collaborations and if family responsibilities constitute a glass fence – a gendered obstacle that keeps women from this engagement. Using an international data set, we find that women engage less in international collaborations than men, and that complex gendered patterns exist regarding the impacts of partner employment status and children. Both men and women benefit from having an academic partner, although men benefit more. Partner employment status matters more than children in certain family arrangements, suggesting that the former constitutes a glass fence, potentially impacting women's access to cutting-edge international knowledge production and elite academic positions.

Keywords:

• gender
• international research collaboration
• parenting
• partner status
• academia

Acknowledgements

The authors gratefully acknowledge Martin Finkelstein, professor of Higher Education at Seton Hall University, for providing them with a copy of the CAP 2007 international data set. Many thanks, as well, to Michael Handel, Alan Clayton-Matthews, Julia McQuillan, Christina Falci, Helen Connolly, Agnete Vabo, the Methods Consultants at Ann Arbor, and the participants at a roundtable session at the SWS conference in 2011 for their thoughtful exchange and suggestions.

Disclosure statement

No potential conflict of interest was reported by the authors.

Notes

1. Argentina, Australia, Brazil, Canada, China, Finland, Germany, India, Italy, Japan, Korea, Malaysia, Mexico, the Netherlands, Norway, Portugal, South Africa, Venezuela, the UK, and the USA. While we were not involved in the data collection, we received access and permission to analyse and publish the data.

2. Because of the negligible number of part-time employed partners in academia (N = 132), we combined them with the full-time employed academic partners (N = 1862).

3. Unfortunately, our data set does not allow us to tease out nuances between families that do or do not reflect a model of a heterosexual couple with children. In many societies, families take diverse forms, including lesbian, gay, bisexual, and transgender (LGBT) parents and children from earlier and current relationships. In some cases, more than two parental figures might be involved, meaning that care for children is divided over different households.