Ensuring Sustainable Evaluation: How to Improve Quality of Evaluating Grant Proposals?

Abstract

The principle of sustainable development is an obligation placed on all entities involved in the implementation and delivery of the structural funds made available not only by the European Commission but also by grant donors from all over the world. For this reason, when applying for a grant, proposals need to demonstrate the positive or neutral impact of the project on sustainable development. To be able to select projects that will ensure sustainability, we need to ensure the effective evaluation of the proposals. The process of their evaluation should be objective, unbiased and transparent. However, current processes have several limitations. The process by which grants are awarded and proposals evaluated has come under increasing scrutiny, with a particular focus on the selection of reviewers, fallibility of their assessments, the randomness of assessments and the low level of common agreement. In our studies, we demonstrated how some of those limitations may be overcome. Our topic of interest is the work of reviewers/experts who evaluate scientific grant proposals. We analyse data coming from two prominent scientific national grant foundations, which differ in terms of expert’s selection procedure. We discuss the problems associated with both procedures (rating style of the reviewers, lack of calibration and serial position effect) and present potential solutions to prevent them. We conclude that, to increase the unbiasedness and fairness of the evaluation process, reviewers’ work should be analysed. We also suggest that, within a certain panel, all grant proposals should be evaluated by the same set of reviewers, which would help to eliminate the distorting influence of the selection of a very severe or very lenient expert. Such effective assessment and moderation of the process would help ensure the quality and sustainability of evaluations.

Cite this work

No public copy is stored for this record, so only the metadata is cited.

Stored only when both papers are in the corpus — 4 citations exist upstream.

Papers that cite the same references — the relational fallback while embeddings are still backfilling.