Il problema metodologico del referaggio: cosa vuol dire «valutare» un saggio

The paper proposes first a «guide» to the practice of refereeing, presenting some prominent examples of «how» are evaluated the products submitted for publication in journals. The examples allow understanding what are the procedures used and the profiles on which assessment is required (originality, structure, data correctness, accuracy of references, etc.). Are then presented some of the most significant problems that the vast literature on peer review has highlighted, in particular: how effective it is, that is able to detect misconduct (falsification of data, plagiarism, etc.), how much is fair, that is able to avoid the creation of systematic discrimination against certain groups of researchers or non conventional shapes and innovative research, how efficient, ie not responsible for slowing the spread of new research and overloading the best researchers.