What’s wrong with research culture?

UK research, including chemistry, punches above its weight. A 2017 report commissioned by government found the UK has 4.1% of the world’s researchers but produces 15% of the world’s most highly cited articles. This might then indicate a healthy and flourishing research community, but many say this output comes with too high a human cost. They point to a lack of diversity in academic appointments, a leaky pipeline that sees female researchers leaving the sector before they reach senior roles, and alarmingly high levels of bullying and harassment. In addition, early career researchers face years of temporary contracts and insecurity, with only 10% likely to gain permanent academic positions.

‘Previously we’ve tried to tackle most of these issues in isolation,’ explains Karen Stroobants, a research leader at the policy research organisation RAND Europe. But it’s possible to see them collectively as the results of problems with the ‘research culture’. ‘[The concept] has recognized that there are really strong links between all these areas and there’s an interconnectedness to the system that we don’t fully capture when we try to address an issue in isolation.’ Stroobants was previously science policy lead at the Royal Society of Chemistry and part of the Royal Society’s research culture programme team. From 2017, the Royal Society started looking into a number of problems that together add up to an unhealthy research culture in UK universities, which often impacts laboratory-based subjects such as chemistry more than others.

The reality of the very competitive environment is you get collateral damage

So where did it all go wrong? According to Stroobants, one of the central issues is hyper-competition. ‘Competition can be good, but I think we are now in an area where there is competition to such extent that it is causing negative consequences,’ she explains. Mark Miodownik, a materials scientist at University College London, agrees. ‘The number of people who want to have a career in science has gone up exponentially in the last few decades.’ In recent years he has been interested in how we can improve research culture and chaired a 2018 Royal Society conference on this issue. ‘There is no way I would get a job now on the CV I produced to get my first job and I think if you talk to most academic my age, they will say the same thing,’ he concedes. But the problem isn’t restricted to hiring. ‘Once you’re in the department, if you’re not producing high quality research papers and getting grants, you’re under a lot of pressure,’ he explains. ‘The reality of the very competitive environment is you get collateral damage.’

Competition has always been high, according to Paul Walton, a chemist at the University of York. ‘And rightly so… I don’t think we should shrink from that.’ But what arises from this can be a sort of ‘machismo’, he explains, an aggressive approach to the competition that can create a negative culture. This becomes offputting to many people, and ultimately impacts the diversity of research staff – an issue Walton has worked to improve for many years.

Measure for measurement’s sake

Where competition becomes unhealthy, it is often down to the use of metrics – quantitative measurements designed to evaluate research outputs – in judging academic staff. This is well expressed by Goodhart’s law: ‘when a measure becomes a target, it ceases to be a good measure’, itself first expressed in those terms in a paper looking at excessive auditing of UK universities.

‘Academia is probably the most measured profession in the UK and that constant measuring, it’s distorted what we do,’ says Walton. Miodownik agrees that the pressure from government in exercises such as the Research Excellent Framework, which allots departmental funding based partly on metrics, has made universities more focused on a narrow set of ‘quality indicators’ and less likely to ‘take a punt on someone who they think has got promise’. This then creates a ‘very narrow definition of what success looks like’, argues Stroobants.

They found that a big barrier to open science was the individualistic incentive structure

While metrics can measure a variety of factors, the most common and contentious indicators assess the quality of a researcher’s publications, based on the impact factor of the journal the paper is published in, or the newer h-index, which gives a longer-term researcher-level metric that measures both their productivity and citations. These kinds of metrics have been around since the 1980s and were originally designed for use at a government level to assess large groups. ‘Things changed in the 2000s when the online versions of the bibliometrics became available and then those desktop bibliometrics began to be used by the heads of departments without an understanding of what those numbers meant,’ says Ismael Rafols, an expert in research metrics from the University of Leiden in the Netherlands.

The upshot of the ‘metrics culture’ is that publishing becomes an end in itself rather than a means of communicating results to advance the creation of further knowledge. ‘By focusing scientific activity on publishing papers, other important scientific activities, such as organising seminars in the department, participating in peer review in the scholarly community, organising seminars for learned societies, training and teaching, all these activities become less valuable,’ says Rafols.

The focus on publications also discourages moves towards sharing data, something funders want to encourage. ‘They found that a big barrier to open science was the individualistic incentive structure,’ says Rafol. ‘It was difficult to get people to share, to be more transparent, to publish in open access journals, if researchers are in a competition that is related to metrics.’

There are also indications that the current metric environment could even change the type of science being carried out. This is difficult to prove because of the myriad factors involved, but Rafols says, ‘there are good reasons to believe that metrics have accentuated some directions. So for example, they might have exacerbated the tendency to focus on hot topics, or topics that are well funded to the detriment of wider variety.’ The incentive to move into areas of research likely to be published in top journals may also work against some forms of applied research, instrumentation research and methodological developments.

In the last decade there has been some community push-back against the overuse of metrics. This includes the Declaration on Research Assessment (Dora), developed in 2012 during the annual meeting of the American Society for Cell Biology in San Francisco to set out standards for metric use, as well as the Leiden manifesto from Rafols and colleagues in 2015. In the UK the sector-wide Forum for Responsible Research Metrics has provided advice to UK funding bodies on the use of metrics in the coming 2021 REF.

But as of July 2021 only 16 UK universities had signed Dora and misuses of metrics continue, including Liverpool University’s intention to use citation metrics in making decisions on redundancies, reported in April this year. This was a first in terms of an explicit policy, says Rafols. ‘In the end they didn’t use bibliometrics, only research income metrics. We know that informally people have used [biblio]metrics to make [these kinds of] decisions.’

The precarious career ladder

To progress to the position of a constantly measured member of the academic world, researchers must negotiate a precarious path, first becoming postdoctoral or early career researchers. ‘Only a very small percentage of PhD students will go on to have a career as a postdoc, and then in turn, only around 10% of postdocs will go on to find permanent positions in academia,’ says Stroobants. Those who choose to continue are almost inevitably looking at years of fixed term contracts. 2018 data from the Higher Education Statistics Agency indicated that nearly 33,000 UK researchers (67% of the total ) were in this position. This is now the default business model for universities – a version of the ‘gig economy’, where only a small elite have any job security.

The effects of this ‘casualisation’ of research roles has been investigated by Nick Megoran, a political geographer at Newcastle University in the UK, who says the practice, common in science subjects, is now creeping into other disciplines. He became interested in the issue in 2017 when Newcastle attempted to implement a new performance review for researchers while also commemorating the 50th anniversary of a visit from Martin Luther King Jr, who spoke on the dignity and respect to which all humans were entitled. ‘At the same time Newcastle was using this [to promote] its brand, we were doing this dire thing that was utterly dehumanising [research staff],’ says Megoran. With colleague Olivia Mason, he decided to explore the impact of short-term contracts and in January 2020 published the report for the University and College Union (UCU). They characterised the system as turning early career researchers into ‘second class citizens’ within the academy.

The impact of this is far-reaching, from an inability to take control of their own research agendas to being unable to plan the most basic parts of their lives. ‘Women can’t choose to have babies because there [may be] no maternity leave, people can’t buy homes because you can’t get a mortgage. All of these things have a very dehumanising effect,’ explains Megoran. He concludes that the current culture makes early career researchers feel ‘invisible’ and has a significant impact on their mental well-being.

Chemistry does seem to suffer from a very large drop-off of female staff

The funding system is clearly now established to work this way, but Megoran says that universities do have a choice. Miodownik points to Europe where universities tend to employ a certain number of fully funded junior position. ‘You can employ the person you think is best for it for five, six years… the position is not going away.’

This early job insecurity is a significant factor in the lack of women in senior academic roles in chemistry – an example of the interlinked nature of these research culture issues. A 2018 RSC report identified funding structures, academic culture and difficulties balancing responsibilities as key barriers blocking women’s progression and retention in research. While 44% of undergraduate chemistry students in the UK are female, only 9% of UK chemistry professors are. ‘Chemistry does seem to suffer from a very large drop-off of female staff, from junior to senior grades – I think only clinical medicine is worse,’ says Walton. ‘Diversity has struggled to make headway in academia… the UK has made progress, but it remains difficult to get it fully integrated.’

‘I don’t think it’s a level playing field,’ says Miodownik. ‘You can see a certain type of person gets the [academic] job, and it hasn’t really shifted, and that should ring alarm bells, for everyone.’ He argues the current lack of diversity in our researcher workforce ultimately damages science itself. ‘If we’re not getting the best people in the door, then science isn’t going to get the best.’ Diversity is likely to prevent group-think and provide different perspectives on the sort of science needed to address societal needs; ‘Obviously the questions that get addressed are biased by the people who do them, and so we have to reflect society,’ adds Miodownik. ‘We need people who are not part of a clique.’

Bullying and harassment

Women are often disproportionately affected by harassment and bullying in academia, and the resulting high levels of mental-health problems, particularly in early career researchers. A 2017 survey by the UCU found that one in four UK respondents said they had been bullied by a colleague – significantly more than the one in seven among EU respondents. A study of PhD students the same year found that 32% were at risk of having or developing depression or other mental health problems. Sexual harassment is also rife in academic departments and research shows it very often goes unreported, which in 2019 led the RSC to launch the first harassment helpline for chemists.

A US study published in 2020 suggested that respondents in the life and physical sciences were 1.7 times more likely to keep quiet when harassed than respondents in other disciplines and many link this to a culture in science that sees large power imbalances between supervisors and their research group members. ‘Young [researchers] are often dependent for references and for career progression on one supervisor who can make or break their future career,’ says Stroobants. Given the intense competition for academic jobs and the sink or swim nature of those first few years, it’s not surprising that many feel powerless to speak up.

Issues of bullying and harassment are not unique to the academic workplace, says Stoobants. ‘[But] there is a context within academia that makes it easier to get away with some of these behaviours.’ ‘The rather closed and insular nature of some research efforts, with projects defined [solely] by the supervisor, have been somewhat vulnerable to adverse cultural effects,’ says Walton, which leads to problems going unnoticed for a very long time ‘until something snaps’. This is compounded by many universities’ unwillingness or inability to intervene, particularly against established academics who bring in large amounts of funding.

Some people may fail to see whether a line is being crossed

Ultimately this means often there are few consequences for bad behaviour, and an attitude that ‘academic excellence’ trumps everything else. ‘If someone is really good at doing research, but on the way, they hurt people and maybe even cause mental health issues for other people, that is not excellence to me,’ asserts Stroobants, ‘But what constitutes excellence is a discussion that’s ongoing and I don’t think there’s actually agreement on that in the sector yet.’

Publish or perfidy?

The combination of high pressure to produce results and an imbalance of power means recent years have seen increased reports of research fraud and dishonesty. It is difficult to know the scale of these behaviours, which can range from falsification of data to smaller corner-cutting or unintentional errors that don’t always get picked up during peer review. Serious misconduct is still rare, but according to one chemistry lecturer who wished to remain anonymous ‘some of these behaviours are already quite ingrained in the community, so some people may fail to see whether a line is being crossed… they think that’s just the way it works’.

Postdocs are particularly vulnerable to unethical behaviour of supervisors. Megoran recalls one interview with a researcher on a temporary contract who was being pressured to add co-authors to her papers who had made no contribution; ‘She felt that she couldn’t resist because if she made a complaint she would not get her contract renewed.’ A group of postdocs at the University of Cambridge have even set up a campaign ‘Bullied into bad science’ to tackle some of these issues, particularly allowing researchers to chose to publish in open access journals.

Barring a very few exceptions, the community doesn’t police itself 

The system presents little deterrent to unethical behaviour and misconduct. Unlike the US, the UK has no independent regulatory authority and the current concordat, to which most universities are signatories, is toothless, with many failing to comply with the basic annual reporting requirements. Without a dedicated body, policing is left to universities, funders or journals, whose actions some think are inadequate. UKRI has recently announced the launch of a committee on research integrity, which will be consulting on its priorities this autumn.

The reaction of the community in general can also be troubling when unethical behaviour is uncovered, particularly by more senior academics. Many would rather sweep it under the carpet and there are few communal sanctions on offenders. ‘Barring a very few exceptions, the community doesn’t police [itself] and it doesn’t act,’ says the same anonymous chemist, who puts this down to a general fear of repercussions. ‘We’re all reviewed anonymously for grants and papers all the time and you could very easily make an enemy of someone, so if you can avoid that, then you probably would.’ This suggests the current culture has no ability to weed out those prepared to treat colleagues and students badly. ‘It worries me that some academics are put on a pedestal, despite the way they behave.’

A brighter future

The UK academic sector is clearly still able to produce excellent research, but the testimonies of many within it paint a picture of a problematic culture. Many issues are common to the whole of academia, says Walton, but ‘one culture which is associated with some areas of chemistry is a long hours culture’, which presents even greater opportunities for dysfunctional practices to develop unchecked.

There are signs that the issue is being taken more seriously. In 2020 the new UKRI chief executive Ottoline Leyser spoke in support of change, particularly in relation to greater diversity of research staff: ‘We must reshape the system so that it genuinely values and supports difference.’ Miodownik says change is beginning at UCL. ‘There has been a recognition across the board that there is a problem, and that we need to do something about it.’

Stroobants and former colleagues have set up an online platform, MetisTalk (named after the Greek goddess of deep thought) to provide a forum to discuss the issues that impact early career researchers. ‘I’m quite positive in terms of the amount of discussion that is going on… a lot of people are engaged in the debate, and a lot of people are starting to realise this is important,’ she says.

We need to dismantle quite a lot of the hierarchy

Walton also points out that there have been some positive changes in academia in the last few decades. ‘Health and safety [standards] particularly, have improved enormously during that time, and we should all be grateful for that.’ He says there is also more accountability to the wider world, largely because students now pay fees and are therefore more engaged in the quality of the teaching they receive. Miodownik, a former Royal Institution Christmas lecturer, also thinks that it’s now widely accepted that academics should be involved in public engagement activities. ‘That has shifted in my lifetime… it feels like we have really made progress.’

But there is still resistance to some of the fundamental changes that might make a difference. ‘For instance, very few promotion procedures have changed, the pecking order of research journals remains the same… the way that conferences are held is still the same. Things that really make a difference and motivate people haven’t shifted much,’ says Walton. ‘We need to dismantle quite a lot of the hierarchy,’ suggests Midownik. In his view, there needs to be a flatter structure, rather than the current system of research groups headed by one academic, with PhD students could be part of wider doctoral training courses managed by several people. But not all change is welcomed. ‘I’ve actually raised this in meetings explicitly and I do not get any support for that idea.’

Part of the problem could be that the current system tends to favour senior established academics who have reached the top. Megoran says he knows some of his colleagues do already fight for better contracts for more junior staff. ‘Things will get better if academics take initiative… if we do that together as a body of academics committed to scholarship and treating each other well, we can make a difference. If we simply decide to bunker down in our own little silos and advance our own careers, then we will not make a difference; the choices is ours.’

If these research culture issues are not tackled, it is likely to be to the detriment of the whole research community – and science itself. ‘We run the risk of sticking with an old model of academia, which doesn’t embrace the possibilities of more progressive working methods, particularly at the [subject] interfaces,’ says Walton. ‘Academia will start to miss out on the very best people, as they see working practices elsewhere are far more attractive.’

Stroobants says it’s time to start looking closely at the behaviours our current methods of assessment and evaluation are incentivising and contrast that against what we really value in our research system. But she warns that change won’t come tomorrow. ‘We need to be realistic about the fact that it’s a long game.’

Rachel Brazil is a science writer based in London, UK