Shababi, Hooman (2026) From Metrics to Reflexive Evaluation: Governing Science, Technology and Innovation Policy Without Losing Public Value. Working Paper. fteval Preprints. (Submitted)
Preview |
Text (PDF file)
5901_fteval_reflexive_evaluation_pre-print.pdf Download (206kB) | Preview |
|
Text (Word document file)
5901_fteval_reflexive_evaluation_pre-print.docx Download (43kB) |
Abstract (english)
Science, technology and innovation (STI) policy increasingly governs through indicators, rankings, benchmarks, evaluation rubrics and funding formulas. These instruments are useful because they make dispersed activities visible and support comparison, coordination and resource allocation. Yet they can also narrow policy attention, encourage strategic adaptation and displace the public, epistemic or social purposes that evaluation is intended to serve. This praxis-oriented article develops a framework for reflexive evaluation capacity. It argues that evaluation should not be treated as a neutral measurement stage at the end of a policy cycle, but as a socio-technical arrangement that actively shapes what institutions do and what policymakers can see. The article identifies five linked processes: valuation, commensuration, metricization, economization and performativity. It then proposes five design principles for policy evaluation: plural valuation, explicit theory of change, contestable indicators, protected spaces for qualitative evidence, and periodic review of metric effects. A practical implementation sequence is offered for ministries, funding agencies and research-performing organisations. The framework aims to help STI policy actors retain the coordinating benefits of metrics while reducing proxy displacement, gaming and the invisibility of public value.
| Subjects: | fteval Preprints |
|---|---|
| Number of Pages: | 6 |
| Identification Number: | 10.22163/fteval.2026.768 |
| Date Deposited: | 09 Oct 2026 06:25 |
| URI: | https://repository.fteval.at/id/eprint/849 |
Actions (login required)
![]() |
View Item |


