Future of Research Boston 2015: Can data collection about the scientific enterprise help advocate for change?

Michael Teitelbaum discusses current experiments in data collection on the biomedical workforce. Photograph by Jessica Polka

Michael Teitelbaum discusses current experiments in data collection on the biomedical workforce. Photograph by Jessica Polka

On October 22-24th 2015, the 2nd Boston Future of Research Symposium was held at Boston University. A meeting organized entirely by early career researchers, the goal was to discuss what data should be collected about the scientific workforce: in particular, data about the numbers and demographics of people within the workforce; where they come from; and what their career outcomes are. The meeting also sought to discuss how this data could be collected, and whether this data could be used to advocate for change.

 

In discussions about the biomedical workforce that have occurred over the past two years, it has become clear that many recommendations, and criticisms of those recommendations, are being made with very little data to support or refute them. The most striking example is that the number of postdocs in the U.S. is unknown (between ~39,000 and ~80,000 depending on various estimates). The NIH Biomedical Research Workforce Working Group Report and the National Academies Report, “The Postdoctoral Experience Revisited”, have estimated that the number suggested by the recurring NSF Survey of Graduate Students and Postdoctorates may be off by a factor of 2. In this climate, how can we recommend change or reform to how science is structured, if we don’t know how many postdocs we have and where they go? Can we therefore collect data about the workforce and use that to suggest better models for how the scientific enterprise should continue?

 

The conference consisted of the following panels (you can watch all of the sessions here, and view slides and materials as they are released here):

 

  • postdocs from Future of Research Boston, NYU and Chicago, and the National Postdoctoral Association, who had organized their own meetings and advocacy efforts;
  • postdoctoral office representatives from Boston institutions, presenting data about their postdoctoral researchers to discuss data that they have available and difficulties they experience in data collection;
  • biomedical academics and labor economists presenting data on the workforce and arguments for and against changes to the structure of science;
  • representatives from various career paths, following a presentation from labor economist Misty Heggeness on trainee career outcomes, discussing careers for trainees;
  • representatives from various publishing models (The Winnower, Faculty of 1000, Cell Press and Nature Publishing Group) discussing the role of early career researchers in publishing; and
  • a discussion about the role of data collection in advocating for and creating a diverse research enterprise.

 

Our group in Boston has been keen to get input on this issue from outside of the biomedical workforce, in particular talking to labor economists who study science. Our keynote speaker on the issue was Paula Stephan from Georgia State University, author of “How Economics Shapes Science”, who presented data on “The Economics of the Postdoctoral Position”. She then participated in a panel with Michael Teitelbaum of Harvard Law School (who discussed several recent experiments about the workforce), Melanie Sinche from Jackson Labs (who recently surveyed postdoctoral researchers), Eve Marder from Brandeis University (who wrote this piece in eLife about graduate student populations), and Jonathan Dinman from University of Maryland College Park. This panel discussed possible reforms to the labor force and, in particular, the economics of benefits and living wages for early career researchers, and generated a very lively—sometimes passionate—discussion about how we should be structuring our workforce. One of the most controversial discussion points was the discussion of slightly reducing (or indeed, simply halting the expansion) of graduate student numbers, which Paula Stephan argued for, but Eve Marder was against. When asked who in the research community should be leading efforts for change, Paula Stephan stated that this should be the National Institutes of Health (NIH) and the National Science Foundation (NSF).

 

 

The academic data panel was followed by a discussion about careers and where graduate students and postdocs go, initiated by Misty Heggeness, labor economist on the biomedical workforce at NIH (our unsuccessful efforts to find representatives from non-academic fields which hire PhDs, but also have data on how many they hire illustrated the fact that this data is not readily available). Misty Heggeness has started studying this problem only recently for the NIH and efforts to study the career outcomes of trainees are still very rare and confined to only a few institutions. A discussion about careers and perceptions, on what various fields actually need/desire from candidates, followed with Brian Plosky from Cell Press, Chris Pickett from the American Society for Biochemistry and Molecular Biology (ASBMB), Tyler Ford from Addgene, and Sarah Cardozo-Duncan, a career strategist in the Boston area. The panel had more in the way of advice than data, which we find interesting and a point of future focus. However, although there are increasing recommendations to train PhDs for non-academic careers, there seems to be little to no data about the number of jobs available to justify the scientific enterprise producing so many “trainees”.

 

My own personal highlight was the panel discussing data collection in improving diversity of the workforce. In contrast to discussions about data collection assisting with advocacy in the rest of the symposium, this panel gave me pause for thought on how collection of data on diversity needs better metrics than what we have now (and Joan Reede from Harvard Medical School challenged us to think about what those metrics could be). Alberto Roca from MinorityPostdoc.org pointed out that collection of data can become a distraction, and in some cases no amount of data can replace simple advocacy for equal rights, and reducing obvious biases in hiring and mentoring. In addition, Moon Duchin from Tufts cautioned us to think about broken mentoring of underrepresented minorities in science—in particular, that overpraise and a lack of constructive criticism are often given to minority mentees and are ultimately more damaging, though perhaps well-meaning in intent. These issues may become particularly acute when departments or mentors may only worried about metrics or the ability to get funding, instead of whether people are able to fit into academic programs. Then students may be disadvantaged because they aren’t getting enough, or the right kind of, mentoring. This panel was therefore incredibly helpful in getting participants to try to think more intelligently about advocating for greater diversity in the research workforce.

 

Boston attendees at the ASBMB/FOR Hack Day, photo courtesy of Chris Pickett.

Boston attendees at the ASBMB/FOR Hack Day, photo courtesy of Chris Pickett.

In discussing what should be collected, we also require actionable output and with ASBMB we held a Hack Day event that included groups in Boston and at Washington University in St Louis, MO. Sitting down to tackle various problems in groups, filled with questions that we had developed over the course of the meeting, many exciting products were developed that are currently being judged with the potential for prizes from ASBMB, and we hope to release more details about these soon!

 

We are looking through all our materials to again create output for the meeting in a similar manner to last year, to try to advocate further for more evidence-based policy recommendations in changing scientific research, and also to generate experiments to test hypotheses about the way science is being done. It is my hope that we will be able to advocate for changes to the research system based on the vast amount of data presented to us; to discuss what further changes should be made in data collection to assist further advocacy; and to focus on areas requiring advocacy efforts foremost without being distracted by data collection, in our future efforts towards creating a sustainable research enterprise.

 

Do you have any ideas on what data we should be collecting, and what the best way is to be thinking about data in advocating for change in the biomedical workforce? Please post them below!

About the Author:


Gary McDowell is Executive Director of The Future of Research, Inc. (http://futureofresearch.org/), a nonprofit organization seeking to champion, engage and empower early career researchers with evidence-based resources to help them make improvements to the research enterprise. He is a COMPASS alumnus.