Exonerations in the United States, 1989 Through 2003: The statistics

This is a very exhaustive study of exonerations, and we will be bringing you some highlights from it over the next few weeks. Click here if you want to read the entire study yourself.

The study's authors are all associated with the University of Michigan:

Samuel R. Gross, Thomas & Mabel Long Professor of Law
Kristen Jacoby, J.D. candidate, May 2005
Daniel J. Matheson, J.D. candidate, May 2004
Nicholas Montgomery, Ph.D. candidate, Department of Economics and Ford School of Public Policy
Sujata Patil, Ph.D. candidate in biostatistics, School of Public Health

Here are a few statistics from the study, which covered a period of 15 years (Note: the exonerations identified do not include 135 innocent defendants in the Tulia and Rampart mass exonerations or the 70 childcare sex abuse cases):

• 328 exonerations, 316 men and 12 women
• 145 cleared by DNA, 183 by other sorts of evidence
• 199 (61%) were murder cases; 73 (22%) were sentenced to death
• 120 (37%) were rape cases
• 6 were for other crimes of violence
• 3 were for drug or property crimes
• 88% of the rape cases were exonerated by DNA
• 20% of murder cases were exonerated by DNA, and almost all of them also involved a rape
• Defendants convicted of murder make up about 13% of the American prison population, but 61% of all exonerations, and 87% of non-DNA exonerations
• Death row inmates make up 1/4 of 1% of the American prison population, but 22% of exonerations
• The 4 leading states for exonerations are: Illinois (54), New York (35), Texas (28), and California (22)
• Eyewitness misidentification is the primary reason for false convictions in rape cases; particularly cross-racial misidentifications
• The leading cause of false convictions in murder exonerations was perjury: by the real killers, by supposed participants or eyewitnesses, by jailhouse snitches and other police informants, and by police officers and state forensic scientists
• False confessions also played a large role in false murder convictions: 44% of juvenile exonorees and 13% of adult exonorees falsely confessed; 69% of exonorees who were mentally ill or mentally retarded falsely confessed, compared to 11% of exonorees without any known mental disabilities falsely confessed
• 90% of all juvenile exonorees are Black or Hispanic

Regarding the disparate number of death row exonerations, the study considers two possible explanations:

1) False convictions are not more likely to occur in murder and capital cases but are more likely to be discovered in capital cases because of the care taken to review those cases.

2) False convictions are more likely to occur in murder cases, and far more likely in death penalty cases.