A leading programme to train PhD students in the UK has come under fire after many top-rated centres were told that they will not receive any further money. According to a Physics World investigation, the Centres for Doctoral Training (CDT) programme — funded by the UK’s Engineering and Physical Sciences Research Council (EPSRC) — will see around two-thirds of its top-performing hubs being axed. Researchers running centres that have not been renewed say they have also not been told why — despite in some cases receiving top marks during the selection process.
The CDTs are funded by each of the major research councils in the UK that belong to UK Research and Innovation (UKRI). Designed to give students a broader perspective than a traditional UK PhD, each CDT is based at either a single university or shared across multiple institutions. Around 15 students in each “cohort” are funded in a specific research area over a three- or four-year period. CDT students begin taking courses and may carry out two short research projects, which can also be done in industry, before choosing which to continue for a PhD project.
EPSRC’s CDT programme began in 2009 when it funded 59 centres over a four-year period. Following a positive evaluation carried out by EPSRC in 2011, the funding body’s council decided to renew the programme and in 2013 expanded the number of CDTs to 115 based at 40 universities, which led to over 7000 students being trained. A total of £950m was spend on this second round, of which £500m came from EPSRC and £450m from industry partners that included firms such as BP, IBM and Rolls Royce. From 2011, the CDT programme has also been adopted by other UK funding councils such as the Economic and Social Research Council and the Medical Research Council.
We were dismayed to find that we failed with a CDT proposition that had a successful mid-term review score as well as a very strong peer review of the proposal
Paul Conway
In 2017, EPSRC carried out a mid-term review of its 115 centres, which was chaired by Mark Smith, a physicist who is vice-chancellor of Lancaster University. When announcing the results in August that year, EPSRC noted that the CDT scheme was bringing “together diverse areas of expertise to equip engineers and scientists with the skills, knowledge and confidence to tackle current and future challenges”, adding that it is “setting the gold standard for cohort-based doctoral training in the UK”.
However, EPSRC did not initially release the ratings it gave each centre, only doing so following a freedom of information request made last year. That request revealed that EPSRC gave each CDT a score between 1 and 4, with 44 centres receiving a 4, meaning they were rated “good”. However, nearly two-thirds of CDTs were given a score of 3 (“good/satisfactory”) or lower, with 19 CDTs receiving the bottom mark (1), which resulted in the CDT’s principal investigator being called in for an interview to consider whether it would be continued.
Early last year, EPSRC then announced a call for proposals for CDTs to be funded between 2019 and 2024. On 4 February, it revealed that 75 centres based at 31 universities had been successful, which will result in around 4600 students being trained. With a total fund of £830m, EPSRC called the new programme “one of the UK’s most significant investments in research skills”, but it in fact represents a cut of almost 20% in real terms compared to the 2013 programme. Of the £830m, some £446m will come from EPSRC with rest from industry.
Selection process
According to a Physics World analysis, a total of 80 of the 115 centres funded last time will not be further supported. Only 35 of the 2013 CDTs have been renewed with the remainder being entirely new centres. EPRSC claims, however, that around 40 — or 53% — of 2019 CDTs will “build upon a previous EPSRC CDT investment”. Yet this still means that most universities will see severe cuts to the number of EPSRC-funded CDTs. Imperial College London, for example, lost half of its 12 centres. Meanwhile, some institutions – notably the universities of Exeter, Leicester, Loughborough and Southampton — now have none at all.
Concerns have also been raised about the lack of correlation between the results of the mid-term assessment and CDT renewal. For example, only 16 out of 44 of CDTs rated “good” were retained by EPSRC. Among the 28 good performers that were axed were several at the University of Warwick, which from 2013 had five CDTs (as well as a further CDT co-sponsored with the University of Oxford) funded with £20m. In the mid-term review, four were given top marks and the other rated 3, but only the Mathematics for Real-World Systems (MathSys) CDT (rated 4) was renewed with £4.6m (compared to £3.7m in 2013). Meanwhile, a new CDT — Centre for Doctoral Training in Modelling of Heterogeneous Systems (HetSys) — was awarded £5.5m.
New and existing CDTs were treated equally with the same assessment criteria to ensure a level playing field
EPSRC spokesperson
“We are very pleased and proud to receive funding to continue our successful MathSys CDT and to establish [HetSys],” a spokesperson for Warwick University told Physics World, adding that “a number of” Warwick CDTs had reached the “final-stage interview process” towards selection. He adds that Warwick is committed to supporting “a range of Warwick CDTs” alongside the two EPSRC-supported ones. This, he says, will “provide a strong and widespread doctoral training across the breadth of our sciences in collaboration with our industrial partners”. Yet the spokesperson declined to say why those that were awarded a “good” rating failed to be funded further.
Highly competitive exercise
Another top-rated CDT that will not be funded for new students from 2019 onwards is the CDT in Embedded Intelligence at Loughborough University. As well as receiving a grade 4 in the mid-term review, it also got given “strong” feedback during peer-review in the proposal process. However, Paul Conway, who is director of the CDT, told Physics World that he has not been given any feedback on the result. “We were dismayed to find that we failed with a CDT proposition that had a successful mid-term review score as well as a very strong peer review of the proposal, international endorsement and healthy student recruitment numbers,” he says. Conway adds that his team had been told to expect feedback in mid-to-late January, but that they are still waiting, adding that he has “expressed his concerns” to EPSRC over the running of the process. The peril of proposals
Other discontinued CDTs that have been contacted by Physics World also confirmed that they were not informed why their centre would be not receiving funding. This has caused some to speculate that EPSRC wanted to consolidate the CDT programme to a smaller number of institutions or perhaps fund strategic areas of research at a detriment to those that were performing well. For example, while the CDT in Delivering Quantum Technologies at University College London received a 1 in the mid-term assessment, it was still renewed. The bottom-rated CDT in Sensor Technologies for a Healthy and Sustainable Future at Cambridge University, has been renewed too.
However, EPSRC denies any such assertion. “New and existing CDTs were treated equally with the same assessment criteria to ensure a level playing field for new and existing centres and encourage applications for centres in new areas,” a spokesperson for EPRSC told Physics World. “Decisions on which centres to fund were taken using robust peer review in a highly competitive exercise.” EPSRC says that they will be informing unsuccessful centres “shortly” adding that an “exercise of this scale takes a little while to complete”.
EPSRC also denies that there will be cuts to the new CDT programme claiming that overall support for CDTs has in fact grown given that UKRI will provide an additional £100m for CDTs in artificial intelligence, some of which will “support doctoral training within EPSRC’s remit”. The council also states that additional funding of £212m from “academic partners” will take the spending by industry and university partners beyond 2013’s value of £450m. As to why there are fewer CDTs compared to 2013, EPSRC says that this is mainly due to increased PhD fees and stipends for students.
