Using Open Chance in our selection process

Afield will trial Open Chance Selection (also known as random selection) in its selection process. This means that part of our selection process will be subject to chance: a random draw will choose a longlist of applications for us to read and assess in full: 30 for Nature Communities and 50 for Artistic Research. It is called “open” because it aims to be more transparent, and to provide greater equity in who can access funding, than conventional application processes.

Random selection has a long history - for example in the selection of juries - but is relatively new in grant-making. Used in combination with simplified application forms, there are strong reasons for trying it:

• to increase fairness and transparency so that over time applicants from a range of backgrounds are successful
• to reduce conscious and unconscious bias in the selection processes
• to fund a diversity of projects including experimental and risk-taking projects
• to reduce the time and cost of writing and reading applications, for applicants and for Afield.

Context

Many of the current funding models no longer feel sustainable for applicants or for funders. In the arts, as in many other sectors, there has been a steady collapse in funding in the UK, where applicants are chasing a decreasing pot of money. A new arts fund reportedly spent over a month reviewing over 3000 applications it received (1).

Martin O’Leary from Watershed in Bristol estimates that applications for the organisation’s two winter residencies in 2022/2023 required 400 hours of unpaid work in total by the 100 artists who submitted a two-page proposal. He calculated that the total cost of all this unpaid labour exceeded the value of the grants themselves by more than two and a half times. O’Leary is describing a broken model that applies on a larger scale in arts funding as a whole (2).

In real terms, spending per person on culture by Arts Council England (ACE) fell by 18% between 2009/10 and 2022/23, and local government spending on arts decreased by 55% since 2010. One of the consequences is a creeping normalisation of corporate funding models and a culture of competition and market-based practices. More and more demands are placed on artists for fewer rewards (3).

The problem with selection processes

The design and implementation of selection processes are an important and under-discussed part of the arts economy. Numerous biases creep into selection processes. Firstly, accomplished grant writers, or people who can afford to pay for someone to help them write, or write on their behalf, have an advantage. This can reward grant-writing skills rather than the strength of the idea. Moreover, these applicants are more likely to be middle class, have time, be well-educated, and neurotypical (4). This system is clearly rewarding the wrong things.

Secondly, with most funds heavily over-subscribed, there are huge numbers of fundable ideas - way more than can be supported. When it is hard to make meaningful distinctions between applications, selectors tend to favour ideas that are on trend, or known names who are considered a safe pair of hands (5). It is easier to produce consensus in a selection panel by choosing ideas that are mainstream, or are seen as most likely to ‘succeed’ (6). The same people get funded again and again and an effective system of gate-keeping arises. This functions as a kind of accepted ‘black box’ where numerous possibilities are reduced to a handful of successes, with little transparency about how these outcomes are arrived at.

In some settings these problems can be at least partly addressed through anonymisation, so that reviewers don’t know the identity of the applicant. However this is often not possible in the arts where artists can be recognised by their portfolios (7).

The result is not only the reproduction and amplification of bias through the system, but a form of systemic conservatism. A study by the Research on Research Institute (on academic selection) found that unconventional, imaginative ideas, especially those from early career applicants, are less likely to be favourably peer reviewed (8).

How randomisation can help

Alongside a simple application form, randomisation can help address multiple problems of bias. It narrows the pool without reader bias at the initial stage, ensuring that a proportion of proposals that might be seen as risky or experimental are considered. Unusual or imaginative projects have the same chance of being longlisted as conventional ones, and applicants without a strong track record are not at a disadvantage, helping to ‘maximise’ opportunity (9). Together, this can lead to a more diverse range of projects being longlisted - and ultimately funded (10).

The evidence for how this works in practice, though limited, is encouraging. For example, Jerwood Arts’ 1:1 Fund, which experimented with a random selection tool, found that 81% of applicants were applying to Jerwood for the first time, suggesting it can open funding up to people outside the usual circle (11). In an evaluation of the application process by Jerwood, more than half of respondents (54%) said that randomisation made them 'more' or 'a lot more' interested in applying, as they felt it improved their chances of being selected. About a quarter (24%) said it made no difference at all, and 20% said it made them 'less' or 'a lot less' interested in applying (12).

An important variable for Afield is the relationship between randomness and inclusion. Whether randomisation makes funding more inclusive depends on who applies. Will it help or deter applications from individuals who have been historically disadvantaged in funding? We don’t fully know yet, but there is some hopeful evidence around this. According to the Random Selection: How-To Guide, “random selection has often resulted in higher numbers of applicants overall, including higher numbers of applicants from underrepresented groups. This is due to applicants feeling they have a fair chance of getting through.” (13). Afield hopes that, combined with outreach, and with selection panels that represent the communities that funds are hoping to reach, randomisation at the longlist stage can help improve the diversity of the applicant base and of funded projects.

Afield’s response - using randomisation

Afield will use a random selection tool in conjunction with a simple application form. As a small organisation with three part-time staff, we don't have the capacity to give full consideration to more than 50 applications for the arts.

An outline of our selection process:

• Targeted marketing to reach historically underfunded groups.
• A simple form where applicants can use bullet points to answer.
• We will use a randomisation tool to choose 50 applications for the arts and 30 for Nature Communities.
• Eligibility will be checked and we will re-draw if any applications are ineligible so that we give full consideration to 50 or 30 applications, depending on the fund.
• The applications will be assessed internally against the published selection criteria. Unsuccessful applicants will be given feedback in writing.
• A shortlist of applicants will be paid £300 to answer further questions.
• An external panel, representing the communities we hope to fund, will select projects for funding.
• Any applications not selected can re-apply the next year with the same or a modified application.

Randomisation is not a panacea, and some candidates will feel deterred by the possibility that their application won’t be read. It is also only as good as the pool it is drawing from.

What we don’t know yet

Detailed descriptions of the implementation of such processes and of their effects are usually kept private and not visible to those outside of organisations.

At Afield we hope to introduce some transparency by collecting and publishing demographic data relating to overall submissions, those selected randomly, those shortlisted, and successful applicants. We are especially interested in monitoring the relationship of randomisation to diversity and inclusion.

This is not a standalone experiment but will take place in relation to the other elements of the application process, which are all designed to reduce unpaid labour and improve accessibility. We will publish updates on selection process data and would be happy to hear any thoughts or comments you have as a reader or potential applicant. Please contact us at hello@afield.org.uk

Written by: Liz Orton 08.10.26

Notes and references

1. This information was learnt in conversation and can’t be directly attributed.
2. https://www.youtube.com/live/dMJ0VXnOm88
3. There are many other related problems not discussed here - organisations passing costs on to artists themselves, asking for payment to submit a proposal, to pay for their work to be exhibited or pay for the costs of a residency.
4. Random Selection: A How To Guide. Martin OLeary and Victoria Tillotson, Pervasive Media Studio, 2023.
5. In academic settings, there is increasing evidence that conscious and unconscious bias related to applicants’ affiliation, ethnicity, age, previous funding success, research area and gender, affect objectivity in selection processes (Ginter et al., 2011; Hoppe et al., 2019; Li, 2017; Murray at al., 2016; Severin et al., 2020; McAllister et al., 2015). These risks increase when reviewers have limited time to dedicate to evaluations.
6. These articles discuss this in relation to academic panels. ‘Conservatism and risk-taking in peer review: Emerging ERC practices.’ in Research Evaluation, Volume 21, Issue 1, March 2012, Pages 48–60, Luukkonen, 2012. Benda, Wim "The predictive validity of peer review: A selective review of the judgmental forecasting qualities of peers, and implications for innovation in science," International Journal of Forecasting, Vol. 27(1), pages 166-182, G.G. & Engels, Tim C.E, 2011. January.
7. See https://www.ucl.ac.uk/engage/trellis-5-recruit-less-pay-more-and-be-transparent-lessons-ive-learned-trellis for more of a discussion on how these issues have been approached by UCL for its Trellis programme for artists.
8. The Experimental Research Funder’s Handbook, Sandra Bendiscioli, Teo Firpo, Albert Bravo-Biosca, Eszter Czibor, Michele Garfinkel, Tom Stafford, James Wilsdon and Helen Buckley Woods. Research on Research Institute (RoRI), December 2021.
9. Clive Gillman, God Doesn’t Play Dice, but Funding Panels Might. https://substack.com/home/post/p-178186090
10. Why randomise funding? How randomisation can improve the diversity of idea https://www.nesta.org.uk/feature/explorations-initiatives-2020/why-randomise-funding/
11. Random Selection: A How-To Guide. Martin OLeary and Victoria Tillotson, Pervasive Media Studio, 2023.
12. 1:1 Fund Evaluation, Jerwood Arts, 2022.
13. Random Selection: A How-To Guide. Martin OLeary and Victoria Tillotson, Pervasive Media Studio, 2023. See also https://blogs.lse.ac.uk/impactofsocialsciences/2025/04/16/how-randomisation-has-changed-the-british-academys-approach-to-research-funding/