Double-blind peer review and unconventional research: A comparative study of the SNSF Spark Pilot

Abstract In response to concerns that their research funding systems impose unwarranted epistemic restrictions on scientific research and discourage unconventional and risky projects, funders are exploring alternative funding instruments. This paper presents analyses of Spark, a pilot funding instrument of the Swiss National Science Foundation (SNSF) for unconventional and high-risk research that introduced a double-blind evaluation process without review panel discussions. We explore how these programme terms may promote unconventional research by affecting who are applying, their motivations and the assessments of proposed projects. Comparing Spark with a standard SNSF funding scheme we find that the two programmes attracted very different groups of applicants. Anonymized proposals more often inspired non-established and younger researchers to apply, and those inspired to do unconventional and high-risk research were more often funded. Moreover, Spark was effective in avoiding bias against those not holding a professorship or affiliated to a university of applied sciences. Notably, double-blinding was not the sole difference between the two funding schemes, and part of the effects may result from other characteristics of Spark. We discuss implications for how to encourage unconventional research, and the dilemmas of incorporating risk into funding decisions.