Bruce Spencer (Northwestern University - Institute for Policy Research) has posted On Measuring the Balance between Wrongful Convictions and Wrongful Acquittals in Criminal Trials on SSRN. Here is the abstract:
Verdicts in criminal trials can err by convicting an innocent person or acquitting a guilty defendant. Although the ideal error rates are zero, in fact verdicts rest on imperfect evidence and both kinds of errors do occur. Recent work has shown how the rates and relative numbers of both types of errors can be measured statistically both for judge verdicts and for jury verdicts (Spencer 2007). The key is to have multiple ratings of the cases, e.g., also record the judge's belief of the correct verdict in a jury trial or an expert observer's belief of the correct verdict in a non-jury trial. Analysis of a special set of state court cases in 2000-01 from four jurisdictions in a study by the National Center for State Courts (Hannaford-Agor et al 2003) suggested that approximately 7% of the jury verdicts were wrongful convictions and 10% were wrongful acquittals, with corresponding rates of 10% wrongful convictions and 1% wrongful acquittals for the judges' verdicts (Spencer 2007). Those numbers are subject to a number of limitations, not least the non-representative nature of the sample of cases, and those limitations will be reviewed not only for the analysis of the NCSC data but, more importantly, with an eye toward future studies.
Effective data collection and statistical analysis can permit monitoring the actual balance between the two types of error. Blackstone (1825) wrote “that it is better that ten guilty persons escape than that one innocent suffer”. For the NCSC cases studied, the ratios appear to be far from Blackstone's 10:1 - the estimated rates correspond to a ratio of erroneous acquittals to erroneous convictions of less than 2 for juries and less than 0.2 for judges. It is suggested that feedback from the monitoring to the justice system might foster the attainment of societally acceptable tradeoffs between the two kinds of errors.
Very interesting. One interesting normative question arises from asymmetrical distribution of the risk of error. Judge trials, which skew the risk of error towards wrongful conviction, might be thought procedurally unfair for this reason. Another question concerns methodology--impartial observers without independent knowledge of the underlying facts (unfiltered by the trial) have no basis for determining whether the process actually results in correct outcomes. Measuring correctness always requires an independent and neutral factual investigation--something that is missing from this data.