Saturday, 14 April 2018

In Search of the Royal Society Success Rates

Somewhere behind the Palladian grandeur success rates lurk. Possibly (image: Steve Slater, CC BY 2.0).
One of my roles at the University of Kent is secretary to the Research and Innovation Board, which decides on policy and strategy. I’m not a natural bureaucrat, but the paper shuffling, agenda setting and minute writing are compensated for somewhat by the ringside seat I get for the discussions that decide the institutional direction on research and innovation. 

At the most recent meeting, the board looked at success rates for different funders. In readiness for this we had gathered data on how the university compared with funder averages. For the research councils, this was fairly straightforward. Success rates are produced annually for each of the seven councils, via Times Higher. You can see last year’s figures here.

Success rates for the Leverhulme Trust take a little more digging, but the figures can be found in its annual reports. The Wellcome Trust gives more global figures but does provide a comprehensive analysis by gender and age profile, disciplinary distribution and award rate over time. And the British Academy offers all of its figures when its officers give presentations, such as on these slides given to King’s College London.

So it came as a surprise when I could find nothing on the Royal Society’s website about its success rates.

I called and asked if I was missing something. “No, we don’t normally provide those figures,” I was told. The society could provide them, together with figures for my university, but I would have to wait as it was in the middle of a round of financial planning.

A couple of weeks later, the figures arrived. The society gave the success rates for four fellowship schemes: Dorothy Hodgkin, University, Newton International and Newton Advanced. It did indeed provide overall and Kent-specific figures, and even gender ratio, where available. But there was no mention of either the Research Grants or International Exchanges, the two most popular schemes for Kent.

“Unfortunately, I don’t have those numbers,” said my correspondent. I could sense a terseness creeping into the responses. She doubtless rolled her eyes when she saw my name appear in her inbox. But I persevered. “Is there any reason that these aren’t recorded? Or should I approach a different department?”

The reply came from a different department. “The success rate for the Research Grants is dependent on a number of factors, such as the funding available for that round as well as the quality of the applications submitted. I can tell you that for the previous round of awards, the success rate was 19 per cent.”

It was good to have a figure, but I was surprised by the suggestion that the society couldn’t give any historical data because of variance. There probably was a large variance over time, which would affect the overall success rate. But the same would be true for any funder. There was a variation of 30 per cent, for instance, in Leverhulme Visiting Professorships between 2008 and 2015 (28 to 58 per cent), and the success rate for the British Academy Senior Fellowships had both doubled and halved over the five-year period to 2015.

Variation is okay. Variation is all part of the landscape. However, I wondered if the society thought the variation was so marked that the figures were, essentially, unreflective of any long-term trend or mean. “Do you know how much this varies over time?” I asked, like the irritating pedant I am. “And does the International Exchanges scheme have broadly the same kind of success rate?”

“I’m sorry, I do not have the information for the International Exchanges, as they are run by the international team. The success rate is dependent on the amount of funding available, so there is some variation each year. This is why we tend not to publish success rates.”

It seemed an odd response, but it was clearly the party line. Success rates varied, and therefore the society didn’t publish them. “Yes, it doesn’t surprise me that the funding changes from year to year,” I said, and pointed out that the same was true for all funders. In addition, the figures that the society had provided for the fellowships had “a wide (and understandable) variance in both application and award numbers”. I’m sure this was partly based on the amount of funding available.

There was a sense of exasperation in the final reply. “Both the Research Grants and International Exchanges schemes are our largest in terms of volume of applications. Each scheme has multiple rounds per year and each round generates an incredibly high volume of applications. While we have many reports for each round for each of these schemes, we don’t have all the information compiled in the way that we do with the schemes I sent you details of.”

Though exasperated, my correspondent was also apologetic and trying heroically to be helpful. “I appreciate that this is something other funders can provide...we are also trying to provide institutions with as much information as possible. I do understand that we don’t cover enough for you, and I’m sure more institutions, so this is something I will pass on to senior colleagues to consider how we can bridge this gap.”

It did seem odd that, though the society had “many reports for each round for each of these schemes”, they didn’t include the total number of applications or awards—or the percentage of success, which could be obtained simply by dividing one by the other and multiplying by 100. My correspondent finished by suggesting that the Royal Society’s new grant management system, Flexi-Grant, may make the gathering of such data easier.

I do hope so. Having a sense of the viability of different schemes is crucial in helping universities to plan which schemes to target and how best to use the limited time that academics have to write grants. It would also help the society in managing demand for its funding: if applicants know that chances are slim, they will be less likely to apply. And that can only be a good thing for applicant and funder alike.

This article first appeared in Funding Insight in February 2017 and is reproduced with kind permission of Research Professional. For more articles like this, visit

No comments:

Post a Comment