Score contribution per author:
α: calibrated so average coauthorship-adjusted count equals average raw count
It is generally agreed that the conviction of an innocent person (type-I error) should be avoided even at the cost of allowing a certain number of acquittals of criminals (type-II error). The high standard of evidence that is usually required in criminal procedure reflects this principle. Conversely, the established model of optimal deterrence that follows the seminal work of Becker ( 1968 ) shows that the two types of error are equally detrimental in terms of deterrence and thus it prescribes the minimization of the sum of errors with no primacy given to type-I errors over type-II errors. This paper explains that when the costs of punishment are positive, and guilty individuals are, on average, more likely to be found guilty than innocent ones, wrongful convictions are more socially costly than wrongful acquittals. This justifies the bias against wrongful convictions without resorting to any ad hoc assumption about the relative weight of the two errors. Copyright Springer Science+Business Media, LLC 2013