Research quality declines with scientists’ age, study finds

Authors argue this means universities should spend less on senior academics and give promising younger scholars more of a chance

January 20, 2016
Level of quality compass

As researchers get older the quality of their publications drops, a study has found, suggesting that more funding should be given to promising younger scientists.

This latest research contains further bad news for older scholars: it found that they work less well in teams, and after around 12 years into their careers, academics tend to author fewer papers.

These findings add to criticisms that research funding is too heavily concentrated among older scientists, with one Nobel prizewinner telling universities: “put more of your research spend into your under-forties”.

The paper, “How to boost scientific production? A statistical analysis of research funding and other influencing factors”, published earlier this month in Scientometrics, looked at researchers in Canada from 1996 to 2010.

It discovered that career age “negatively affects the quality of published works, which means that as the career of the researchers progresses they tend to produce on average lower quality papers”, as measured by citations.  

In addition, it found that scientists authored more papers per year as they got older, until around 12 years into their career, at which point their output started to drop – although their funding continued to grow.

This indicates that younger researchers with “evidence of a great potential” should be given a chance, rather than continuing to back older faculty “whose scientific performance is already ‘beyond the zenith’”, the paper says.

The study also discovered that the bigger a researcher’s team, the more productive they were and the more citations their papers got. However, this was true only for younger researchers. “As the career age of researchers increases, working in larger teams reduces their productivity and the quality of their papers,” it found.

Co-author Ashkan Ebadi, a postdoctoral associate in the biomedical engineering department of the University of Florida, said the age-related drop in quality and frequency of publications was hard to explain, but that “ambition might be one of the key factors”.

“Younger researchers might be more ambitious in order to prove themselves in the scientific community, or to find a full-time faculty position,” he said.

In October last year, Nobel laureate Brian Schmidt – who has just taken over as vice-chancellor of the Australian National University – complained that institutions direct most of their money to academics over 50, despite the fact that the “apex of productivity” for researchers is usually far earlier.

Responding to this latest study, James Wilsdon, professor of research policy at the University of Sheffield, said that, partly as a result of squeezed funding, UK research councils in recent years had given more of their money to senior principal investigators and less to early career researchers. This “reinforces the danger of creating an imbalance in research funding in the UK and elsewhere”.

But he cautioned against “sliding into ageist generalisations based on citation data” and added that “any move to recast funding patterns based on age would be foolish” and discriminatory. Senior researchers’ time may be spent in crucial management positions, he said, and added: “The research system is more than just a machine for generating citations.”

Register to continue

Why register?

  • Registration is free and only takes a moment
  • Once registered, you can read 3 articles a month
  • Sign up for our newsletter
Please Login or Register to read this article.

Related articles

Reader's comments (5)

Well it is true that my citations dropped a bit when I reached 76. But there's no need to worry since UCL hasn't paid me a penny since I was 68.
I can’t speak for the sciences but a quick search of citation indexes in my areas of expertise in the humanities suggests no such drop in research quality with age. Quite the opposite. As for an age-related drop in frequency of publication, this doesn’t seem at all ‘hard to explain’. If we ‘older scholars’ would only refrain from editing essays to bring them up to standard, mentoring grant applications, supervising dissertations and giving our expert feedback on article and book manuscripts, we’d no doubt have time to churn out a few more publications without regard to building research capacity among younger generations. Need I even mention the vision, planning, people management and basic administration that goes into securing funding for major projects and running them?
The age at which researchers publish their first paper and receive grants has consistently increased. Thus, the findings of this study suggest that the most productive phase of a researcher’s life is further shortened because young researchers begin their career late and are less likely to get grants when competing with seniors in their field. However, it cannot be ruled out that though senior researchers may not be writing as many papers, their experience and mentorship is valuable for young researchers. To resolve this situation, governments and funding bodies should consider merit and impact of research rather than seniority when it comes to giving grants.
This is an interesting paper, but their findings about a relatively early age decline are overstated. Since they do not have information about the chronological age of their sample, they create a variable they refer to as “career age”, which represents the time difference between the date of their first article in the database and the given year of a publication (i.e., between 1996 and 2010). They concluded “Hence according to the curvature of the relationship, although our study covers 15 years from 1996 to 2010, it can be predicted that around 18 years after the start of the work of a NSERC funded researcher his/her scientific productivity starts to decline. Therefore, mid-career NSERC funded researchers seem to be more productive.“ It seems to me that calling this variable “career age” is misleading, because a researcher, who had publications in 1996 could already have been in mid career in 1996 and therefore at the end of his/her career by 2010 (i.e., have a career age of 35 plus). Since their sample is likely to include people, who did not publish every year, it is not only the researchers with a career age (according to their definition) of 14/15, but also researchers with shorter career ages (because they happened not to publish in 1996, or even 1996 and 1997) who are actually at the end of their careers by 2010 (i.e., 40+). And since people have to stop supervising PhD students and also stop getting new grants several years before they finally stop working, empirically working researchers are bound to have a decline in publications during the last years of their careers. Therefore, the slight decline at the end of the authors’ “career age” variable is likely to be due to an overrepresentation of people, who are actually at the end of their career. One of the best recent studies of age and scientific productivity has been published by Gingras et al. (2008) in PlosOne. They studied the publication patterns of a large sample of professors in Quebec (whom they divided into those who were still active and those not active on the basis of whether they were publishing at least one paper in a given year). They found that for active professors there was neither an age-related decline in quantity nor quality of publication. For the active professors, productivity increased to age 50 and then stayed at the same level until age 70. (There were too few older professors to extend the study beyond age 70.)
Really? Just because "a study says" - from one country, and not in any fields that I see mentioned - says this, it must be true? Seriously, in my field of vaccine-related biotechnology, it is simple nonsense. I am 30+ years into a research career, and BECAUSE I am well enough established, I attract more money than most younger folk in my environment do, AND our publication rate and citation rates are high and are not declining. The same can be said of a number of folk around me: our sort of work requires serious application to one's research area over years in order to make an impact, and we do not see a decline such as you speak of until folk are not allowed to take students anymore.