For decades, law enforcement investigations used forensic evidence (i.e. fingerprints, DNA, and ballistics) to include or eliminate suspects. Scientific advances and improvement in forensic analysis have helped to exonerate those convicted of serious crimes. In many cases, forensic biological evidence like DNA was not used. Thus, some individuals convicted sexual assault and homicide may have been eliminated by a forensic analysis. To estimate the rate of such possible wrongful convictions and to identify their predictors, the U.S. Department of Justice's National Institute of Justice (NIJ) funded the Urban Institute (UI) to perform retrospective DNA testing of physical evidence in cases where there was a conviction of a sexual assault or homicide and physical evidence was retained. In June 2012, UI issued a report of the study.

UI reviewed data on Virginia homicides and sexual assaults from 1973 to 1987 (over 534,000 cases). 3,000 of them had retained physical evidence and 2100 of them had an identified suspect. 740 of those cases had at least one suspect convicted of a felony.

Of those, 634 cases with 715 convictions (62 cases had multiples suspects) were eligible for this study based on crime type (homicide, sexual assault) and a conviction.

The DNA evidence was then submitted to a private lab for analysis. "The goal of this DNA testing was to develop a DNA profile from questioned evidence, generally from the crime scene, and compare it to profile of the known persons developed from the original evidence or stored in a database." The profile comparison of questioned evidence would be indeterminate, inculpatory, or exculpatory.

UI researchers reviewed files maintained by Virginia Department of Forensic Science (DFS). All files contained pre-and post-conviction forensic facts, charges, jurisdiction of the crime, and demographic information about the convicted offender, other known suspects, and victims. From this, UI researchers determined if the forensic DNA results will support exoneration, inculpate, or be indeterminate to affect the outcome of the cases. Despite the respective supporting results, other non-forensic facts/characteristics of the case may provide a different conclusion.

Most wrongful convictions were actively claimed and thus detected. Those who have not pursued exoneration remain undetected. Because the Virginia governor ordered DNA testing on all eligible convictions regardless of the guilt or innocence outcome, the DNA testing allowed UI to make predictions that determined the nature of the rightful or wrongful convictions for homicide and sexual assaults between 1973 and 1987 regardless of whether innocence was claimed.

To estimate the rate of wrongful conviction, DNA testing results were classified: When the result supported exoneration, the suspect was eliminated as contributor of DNA testable evidence. With exculpatory but insufficient DNA testing results eliminated the suspect as contributor but other contextual factors did not support exoneration. In some instances, the latter classification was hindered by the lack of data.

Classification could be improved by adding additional court variables (particularly method of conviction (jury/bench trial or guilty plea)), type of defense attorney (court-appointed or retained). Other factors to be added were whether the offender confessed or gave incriminating statements, victim and eyewitness identification, offender's prior record and mental health problems, and results from appeals. This would have allowed UI to test theories put forth in prior studies, particularly about the impact of witness identification, trial type, and confessions on wrongful convictions.

UI interpolated the determinate results of the one-third of convictions and confidentially generalized that all those convictions are an unbiased sample of sexual assaults and homicide in Virginia from 1973 to 1987. To determine the rate of wrongful conviction, a ratio of conviction with exculpatory results, supporting exoneration over the number of convictions was examined.

UI determined the outcome of 715 DNA testing, 65% of the results were indeterminate because they lacked sample reference from the victim and/or the convicted offender. In some cases, the DNA profile only matched the victim or the DNA profile was either unknown or other DNA profile was present.

27 percent of the results were inculpatory because DNA profile matched a convicted offender reference sample but the victim reference sample was not present. In other cases, the convicted offender's DNA profile was present with another profile or with no other profile.

3 percent of the results were exculpatory but insufficient for exoneration because victim reference sample was needed. There was also unknown DNA profile or other DNA profile present.

5 percent of the results were exculpatory and supported exoneration where an unknown DNA profile or other DNA profiles present and the convicted offender had been eliminated.     UI concluded two findings in the convictions in Virginia from 1973 to 1987, where evidence was retained for homicide and sexual assaults cases. First, when the suspect was eliminated as a contributor for a probative item, 5 percent of the convictions support exoneration. Second, 8 to 15 percent of sexual assault convictions and less than 10 percent for homicide cases without sexual assault cases support exoneration.

To preserve confidentiality, the identity of convicted offenders who were excluded from the DNA testing as contributors of questioned evidence. UI concluded that policymakers should be legitimately concerned that justice was not served in those cases where a convicted offender was eliminated.

Source: "Post-Conviction DNA Testing and Wrongful Conviction," Urban Institute, Justice Policy Center, June 2012.