A Matter of Productivity
The Walter and Eliza Hall Institute of Medical Research is a simple place. Our mission is to make discoveries that shape scientific thought and to translate these discoveries into improvements in disease prevention, diagnosis and treatment. To be successful we seek to recruit bright and creative staff and students who are committed to collaborating, sometimes for decades, to crack difficult problems. This is the task to which my Board holds me to account.
At the undergraduate, PhD and postdoctoral level more than 50% of our staff are women. At the laboratory head level the number drops to less than 25 % and at a professorial level, it drops further to 10%. Indeed, when the previous director, Suzanne Cory – at that point the only female professorial member of staff in our 94-year history - retired after 13 years in the job, none of our 20 professorial fellows was female. It did not take a particularly perceptive person to see we had a serious problem. Either we have an even playing field (but make poor decisions when recruiting students and postdocs), hire women that are not creative enough to rise through our ranks and make a long term contribution, or we have created an environment in which talented women do not flourish. The evidence [1, 2] points convincingly to the latter being the case and therefore two questions arise: First, why? Second, what can we do to redress what is, fundamentally, a productivity issue?
In considering this issue, I have tried to describe some of the things that we have done at the Walter and Eliza Hall Institute – I don’t do this to lecture others or to pretend that we have thought of everything or have all the answers – clearly we don’t - but rather to stimulate discussion.
RESEARCH LEADERS THAT EMBRACE THE PROBLEM: For any change in gender equity to occur requires the issue to be embraced by research leaders at board and senior executive level within institutions doing research, as well as in government and philanthropic organisations in charge of funding research. Grass roots, bottom-up action is important, but must be mirrored by a commitment to the problem at an organisational level (e.g. within a strategic plan) and by leadership that can enable change.
A GENDER EQUITY WORKING GROUP WITH A REMIT TO ACT AND REPORT ON PROGRESS: Having leadership embrace change is one thing, but keeping the issue alive in an organisation, over the many years that will be required to cement change, is another. One way to do this is to create a working group who will be charged with implementing the gender equity agenda of an organisation’s strategic plan and who will have a budget that is commensurate with this task. This group should be composed of men and women, senior staff, junior staff and students and should monitor the success (or otherwise) of current gender equity initiatives; suggest refinements to existing programs; and challenge the organisation to be bold and creative.
What do we do when research leaders and their institutions are ambivalent (or worse) about the goal of gender equity? The answer must include a national approach that has both a carrot and stick element – and the UK’s Athena SWAN initiative is a great model. Quoting from their web site: “The Athena SWAN Charter in the UK evolved from work between the Athena Project and the Scientific Women’s Academic Network (SWAN), to advance the representation of women in science, technology, engineering, medicine and mathematics (STEMM).
Athena SWAN has six principles:
▪ “To address gender inequalities requires commitment and action from everyone, at all levels of the organisation
▪ To tackle the unequal representation of women in science requires changing cultures and attitudes across the organisation
▪ The absence of diversity at management and policy-making levels has broad implications which the organisation will examine
▪ The high loss rate of women in science is an urgent concern which the organisation will address
▪ The system of short-term contracts has particularly negative consequences for the retention and progression of women in science, which the organisation recognises
▪ There are both personal and structural obstacles to women making the transition from PhD into a sustainable academic career in science, which require the active consideration of the organisation”
These six principles represent the cornerstone of Athena SWAN; to join, research leaders must indicate that their institution will take action to address these areas.
Over the 10 years since its launch, membership of Athena SWAN now includes all of the UK’s most prestigious universities and research institutions. In addition to providing a whole host of resources to support institutional efforts to redress gender inequity and to provide kudos to those who are leading the way (the carrot), being a member of Athena SWAN and demonstrating a level of commitment to the issue is now a pre-requisite to obtain funding from organisations such as the Department of Health (the stick) [3].
ATHENA SWAN FOR AUSTRALIA
(www.athenaswan.org.uk/)
Why Does Gender Inequity Occur and What Can We Do?
One of the things to keep in mind in considering this issue (and when reading this article) is where the equity problems lie. Although we would love to attract more of the smartest kids in the country (boys and girls) to consider a career in medical research, unlike the physical sciences, girls outnumber boys in their interest, as reflected by enrolment in tertiary courses. Likewise, women outnumber men in embarking on PhDs and postdoctoral fellowships in biomedical research. Clearly the problems do not lie in making biology attractive to girls at school or at the junior level in biomedical science. The challenge we need to overcome lies with women moving from postdoctoral level to faculty and faculty to professorial positions. While there are many initiatives that could make it easier for all women (and men) in health and medical research to balance work and family, when marshalling scarce resources, we should continue to remind ourselves where the challenge currently lies and, at least in the medium term, allocate resources accordingly.
There are many reasons why women do not progress at the same frequency from the postdoctoral level to the lab head level and higher; however, I think that they can be grouped under four broad headings:
- The coincidence of childbearing and career defining years
- The “unpleasantness” of peer-review
- Gender differences to taking on responsibility
- Overt discrimination
Over the next 12 months, in a series of articles on the Women in Science AUSTRALIA blog, I will explore each of these issues and highlight/suggest some simple things that can be done by any research institution or department to redress the problem.
The Coincidence of Childbearing and Career-Defining Years
There are many factors that make juggling a family and progressing a career difficult.
Disenfranchisement of younger researchers and its disproportionate affect on women
A “typical” career path for an academic researcher might be to complete a PhD in their mid to late twenties, embark on a postdoctoral stint in one or two laboratories and then to seek an independent position. When I began my career in the mid 1980s the postdoctoral period was often two to four years after which there was a reasonable chance of beginning a small group of perhaps a student and a research assistant. The trend, internationally, over the last thirty years has been to prolong the postdoctoral period and postpone the scientific coming of age when a biomedical scientist is able to pursue an independent research until the late thirties and early forties. This trend has been disillusioning for all researchers but is particularly invidious for women because the extended postdoctoral period now coincides with the period in which many women wish to start a family. Given a postdoctoral scientist’s productivity is limited to their own scientific output, this creates an awful dilemma – have children and, god forbid, take some time off and have your productivity drop to zero for a period, or postpone having children in the hope that your scientific success as a postdoctoral fellow will allow you to be promoted to laboratory head before your biological clock stops ticking. The situation is a little less dire for a woman who is running her own group, since if she takes maternity leave she will have her students and staff progressing her projects.
APPOINT NEW FACULTY EARLIER AND BASED ON POTENTIAL: Most research organisations have a budget for recruiting and retaining staff. Just as coaches of sporting teams need to make decisions about whether to invest in young players and build for the future or recruit star players from other teams in an attempt to get “instant success”, so research leaders make similar decisions. Should we recruit an established researcher in their 50s or 60s, perhaps with their most productive research years behind them, to a new chair in an attempt to boost our ERA rankings quickly? Or should we promote a postdoctoral researcher in their late 20s or early 30s, who shows great promise, and allow them to run their own group for the first time. Clearly a balance is needed, but the balance desperately needs to be tipped back in favour of youth and promise.
Not all researchers (male or female) will be in a position to be promoted to laboratory head in their late 20s or early 30s; some will take longer to gain the confidence and intellectual independence to run their own research teams; some will make the decision that this is a career track they do not want to go down. For women who choose to have children while being postdoctoral fellows, it is important to allow them to be productive while on maternity leave; this means enabling their projects to progress and, just as importantly (as discussed below), making it as easy as possible for them to be productive when they return to work.
TECHNICAL SUPPORT FOR POSTDOCTORAL FELLOWS ON MATERNITY LEAVE: Provide postdoctoral scientists, with leadership potential, with a research assistant for the period that they are on maternity leave, to allow their project to continue to move forward and hence reduce the risk of “gaps” in their CV.
Dearth of childcare places near work
It is not simply the reduction in productivity during maternity leave that makes it difficult for women to make the transition from postdoctoral fellow to laboratory head. As the primary carers for children in many relationships, as a sector it sometimes seems that we are trying to make it as difficult as possible for women (and men) to balance work and family. In the Parkville area, over the last three years, there has been approximately two billion dollars worth of health and medical research infrastructure built: the major extension to our own building, the Melbourne Brain Centre which houses the Howard Florey Institute, the Peter Doherty Institute and the Victorian Comprehensive Cancer Centre. These facilities, which will house nearly 3000 researchers – more than half of whom are women of childbearing age – have no childcare facilities. ZERO. This, despite the fact that when surveyed, the women who work in these institutions identify childcare close to work as their single biggest need in trying to progress their career while beginning a family [4]. Parkville is not alone in this shame. One suspects that if directors of medical research institutes, deans of medicine or science, deputy vice chancellors of research and vice chancellors had pre-school age children, this situation would have never been allowed to occur.
LINK FUNDING FOR MAJOR BUILDING PROJECTS TO PROVISION OF ON-SITE CHILDCARE: When government and philanthropy contribute the hundreds of millions of dollars needed to build major health and medical research facilities, the funding comes with a myriad of conditions. To prevent a repeat of the current situation in Parkville (and elsewhere), let’s add one more condition to the list … on-site childcare must be included in any new development.
For institutes like the Walter and Eliza Hall Institute, which did not consider on-site childcare a priority during a major re-development, effort must be made to redress this oversight. Whether this is done by creating a childcare centre on site or on near-by land, or reserving places at existing centres is going to depend on the specific environment; however, what is universal is the need to do something, rather than simply seeing childcare as an issue that families need to solve alone.
Making life a little bit easier
Another thing that institutions can do is to recognise that juggling the care of pre-school age children and a demanding career is not simple. Anything that an institution can do to signal to women that it recognises the challenges and will, even in a small way, help them through this period so that they can emerge and enter the next phase of their life and career in a strong position, will be valued. Scheduling all-important meetings between 9.15 am and 3.45 pm is a simple thing that makes life easier for every parent and costs nothing. Other areas worth considering are: financial assistance for regular and/or special childcare; additional time to make a case for contract renewal or tenure; family rooms and purpose-built lactation rooms.
ALLOWANCE FOR CHILDCARE: Recognising that there are economic and time pressures associated with having pre-school age children that add stress to life and can impact scientific creativity and productivity, it is worth considering providing female postdocs (who have an ambition to become independent) and lab heads with pre-school age children with an allowance to help with childcare. For the last few years we have provided up to $15,000 per annum to defray childcare costs. We leave the choice of the child-care arrangement to women and refund costs on provision of receipts. Women can use the allowance to defray the cost of regular childcare or to enable a spouse, parent or carer to travel to a conference with the researcher and their young child. Our mantra is for the researcher to tell us what they need and then we do our best to support them.
ADDITIONAL TIME TO MAKE A CASE FOR REAPPOINTMENT OR PROMOTION: Even with all of the institutional and family support in the world, taking time to have children and caring for them when they are young is a challenge. Recognising that research productivity may suffer, it is worth considering allowing women extra time to make a case for appointment or promotion. In many medical research institutes lab heads are on fixed-term renewable contracts; in the case of the Walter and Eliza Hall Institute our lab head contracts are for five years. We automatically provide female laboratory heads an additional year to make the case for re-appointment or promotion for each child that they have. I am sure some universities have similar policies with respect to tenure. Similar mechanisms should be universally applied to fellowships offered by the National Health and Medical Research Council (NHMRC), the Australian Research Council (ARC) and philanthropy – currently different rules apply to different schemes even within a single funding organization. A more progressive and budget-neutral policy would be to allow a woman (or man) who suspends their for a period and/or who works part-time for a period, to extend the duration of their fellowship support commensurately. For example, and for simplicity’s sake, if a woman with a five-year fellowship works 4 days a week for the entire period while caring for pre-school age children, then allow her to extend the duration of her fellowship for an additional year.
PURPOSE BUILT CARE ROOMS: There are various situations when children of any age cannot attend regular care or school: minor illness for the child or their carer, curriculum days etc. Providing a facility in which researchers can bring children for informal care, allowing parents to meet with staff and students or attend crucial seminars has proven valuable. Likewise, providing a comfortable purpose-built facility for lactation should be seen as compulsory for every workplace. In 2014 it is not acceptable for women to have to express milk in toilets. Would men accept coffee-making facilities next to urinals? I suspect not.
Measuring the important things in peer-review
The final issue I want to raise with respect to the overlap between childbearing years and career defining years is assessment of track record – one of the cornerstones of peer-review. Because women often take time off after having children and, when they do return, often work part-time for a period, there has to be robust means of accounting for these interruptions. There are two considerations in this regard - the first is to consider track record with respect to opportunity. The second is to ensure that the outputs we measure are not simply correlated with the amount of time on the job – what lawyers might term billable hours.
To the credit of the NHMRC, over the last couple of years, it has taken the issue of productivity relative to opportunity seriously and has provided more explicit instructions to peer-review panels; however there is a way to go. Anecdotally, despite peer-review panels receiving instructions, compliance with policy, among reviewers, is variable.
With respect to measures that are akin to billable hours, number of papers may be a poor measure, as might be the number of staff in a laboratory or the number of students or number of invited lectures given. A better measure of research productivity might be the quality and impact of papers, or simply looking for a single profound discovery of major impact. Why peer-review is concerned at all with number of staff or students a researcher might have in their group, or their grant income has always puzzled me; as if there is a magic formula for success! I have been asked at fellowship interviews about the size of my group, as if my seniority or status demands I have a large group. One of my mentors and one of the most senior researchers in the country, Don Metcalf, has seldom had more than three people in his team (including himself) and never had more than four. In the last thirty years he has trained perhaps four PhD students. His argument is that he is most productive when he is focused, working in the lab himself and has time to write-up everything he does for publication. Who am I or any other peer-reviewer to argue? He has always asserted that to train many more students than is necessary to “replace” oneself is an indulgence, a view that has recently come back into currency [5]. Likewise, the preoccupation of measuring a researcher’s success by the amount of funding they accrue also seems perverse; grant income both being an input rather than a research output and an indicator that, at least in part, is akin to “billable hours”.
ASSESSING TRACK RECORD: Whether at an institutional level when considering appointment and promotion or when assessing applications for funding let’s try and keep three principles in mind – (i) discriminate outputs from inputs; (ii) count impact not activity; and (iii) recognise there are many models of doing great science.
References
[1] Bell S. (2009). Women in science in Australia: maximising productivity, diversity and innovation. Canberra: Federation of Australian Scientific & Technological Societies (FASTS).
[2] Moss-Racusin CA, Dovidio JF, Brescoll VL, Graham MJ, Handelsman J. (2012). Science faculty’s subtle gender biases favour male students. Proc Natl Acad Sci USA 109: 16474-9.
[3] Athena SWAN. Charter for women in science, United Kingdom. History and Principles.
[4] WEHI Internal Survey, August 2012.
[5] Alberts B, Kirschner MW, Tilghman S, Varmus H. (2014) Rescuing US biomedical research from its systemic flaws. Proc Natl Acad Sci 111:5773-7