DEFORM Main Findings - A Policy Report

DEFORM was a 26-month project conducted to determine the social and economic costs of harmful research practices (research misconduct) and propose approaches to better identify and model these actions. It is the result of the joint work and fruitful collaboration of INSEEC U’s econometricians and data scientists and the University of Crete’s sociologists, data scientists and criminologists. The project is now finished and has had several interesting results, although some have yet to be confirmed. These results are: 1. A better knowledge of the different approaches and definition of misconduct concepts in research, 2. An enhanced knowledge of the perception of the notion of misconduct in research by the academic research sector and beyond, with a focus on linguistic reluctance with regard to certain qualifications, 3. A clearer vision of the tensions, linguistic and legal misunderstandings that make implementation of transparent and efficient tools for managing indelicate to criminal acts in R&D difficult, 4. A better knowledge of the perception of the gravity of acts committed according to pre-defined profiles, 5. A mapping of critical paths that could lead to research misconduct/fraud, 6. A full costs model of the financial impact of a research misconduct/fraud at macro, meso and micro-levels for an organisation, institution or project, 7. An actuarial model for identifying potential fraud risks by measuring productivity gaps, bearing in mind that — like any risk pricing model — this type of process is based on actuarial assumptions, and thus an estimate of an uncertain variable input into a financial model for the purposes of calculating premiums or benefits, 8. An identification of the standards that could allow a better dissemination of these models, in particular by their appropriation by the actors of the financial analysis, the valuation of assets and the notions of non-value, 9. The participation in the standardisation process (as far as possible knowing that this type of process is standardised and longer than the duration of the project), 10. An approach to the viral dissemination of lessons learned and results obtained during the project through the design of a level 1 certificate integrated into an Executive MBA associated with an app whose maintenance will be free but whose user interface must be improved, which implies working, in the future, on the business model of this type of tool. The present document summarises the most interesting findings and conclusions.