Collaborative Project Between Innocence Project and National Registry of Exonerations Produces Interim Report Reconciling Data Coding Discrepancies
by Casey J. Bastian
The Innocence Project (“IP”) and the National Registry of Exonerations (“NRE”) each keep track of and list wrongful convictions. Each also works to identify the causes of those wrongful convictions and how forensic science-related errors is considered an “influential factor” in many of these injustices. When a study revealed a discrepancy in the contribution coding data between the IP and NRE lists, a five-year reconciliation process commenced. The result was an interim report entitled: “The Contribution of Forensic and Expert Evidence to DNA Exoneration Cases.”
In 2017, researcher Gerald LaPorte (“LaPorte”) published his findings on the relationship of forensic science to wrongful convictions. LaPorte observed that the coding of forensic science as a factor in 342 DNA exoneration cases did not match between the IP and NRE lists. The IP had identified forensic science as having a contributory role in 157 of those cases. Yet the NRE had identified forensic science as a factor in only 133 of the same cases.
LaPorte was raising important questions about the accuracy of the data pertaining to the alleged role forensic science actually plays in contributing to wrongful convictions. The questions in LaPorte’s report concerned “which disciplines were responsible, temporal trends, and the co-occurrence of forensic science with other contributing factors.” LaPorte was essentially arguing that any contributory role is “overstated.” John M. Collins and Jay Jarvis had previously made similar but more explicit allegations concerning this view.
The two organizations assert that the larger overall process will offer a “more accurate accounting of the role of forensic science in DNA exoneration cases.” The goal is to publish a comprehensive report on each of the more than 3,200 exonerations cases identified by the NRE since 1989. One of the first things revealed in the reconciliation process was that LaPorte had actually “understated the problem” of discrepancies in the organizations’ coding processes.
Research into wrongful convictions and the compilation of such lists in the U.S. originated in the 1940s. These lists have served several purposes: they serve as proof that wrongful convictions do occur; researchers are provided the means to count various categories of cases; and measures on issues of interest like contributing factors and their prevalence in wrongful convictions are more readily developed.
Another reason for wrongful conviction list compilation and research is the naming and counting of frequent factors that contribute to, and cause, many wrongful convictions. Over the decades, a “general consensus has emerged around a familiar list of factors.” The IP and NRE each list specific contributing factors. For the IP, it’s eyewitness misidentifications, false confessions or admissions, informants, and misapplication of forensic science. The NRE categories list includes mistaken eyewitness identification, false confession, perjury or false accusation, false or misleading forensic evidence, official misconduct, and inadequate legal defense. These distinctions matter when examining how the coding discrepancies occurred.
The IP originated as a litigation organization purposed to exonerating the wrongfully convicted “using the power of forensic DNA profiling.” The organization enjoyed early success. So much so that the IP joined with the National Institute of Justice (“NIJ”) to collaborate on wrongful conviction list-making and factor-counting. In 1996, the NIJ published “Convicted by Juries, Exonerated by Science,” in which 28 “DNA exonerations” (a person exonerated by post-conviction DNA testing) were first listed.
That publication undermined the skepticism of those who viewed wrongful convictions lists as implausible examples of true innocence. DNA exonerations left little to no room for such challenges. The listing of these DNA exonerations shifted to the book Actual Innocence, written by one of the IP’s co-founders, and eventually to the IP website. For nearly 10 years, this list was considered the “canonical” reference for wrongful convictions in the U.S. Courts, scholars, journalists, and the public relied on this information concerning the risk of producing wrongful convictions.
The problem of only listing DNA exonerations is that it was restrictive. Long before DNA testing of preserved evidence, wrongful convictions were discovered through other means. While highly determinative, DNA testing is only one crucial means for identifying and rectifying miscarriages of justice and freeing innocent people. Exonerations by other means than DNA forensic testing continues still.
To be considered a true DNA exoneration, it requires a very specific set of conditions, and only a small set of wrongful convictions meet them according to IP standards. These include that a perpetrator must leave an “analyzable quantity” of DNA evidence at the scene; there must be consensus that the genetic material was the result of the criminal act and not left for innocent reasons; either inadequate or no testing occurred prior to trial; the person must actually have been convicted; the DNA evidence preserved; and post-conviction testing occurs and produces “probative results.” These stringent criteria mean that a “DNA exoneration” is unlikely in many cases that might still be considered exonerations.
The NRE has been keeping its list since about 1989. In 2012, the NRE began its endeavor to provide “a comprehensive archive of all known exonerations in the United States.” The NRE became responsible for consolidating and rationalizing numerous list-making activities that had occurred during the prior 20-year period. The activities included the works of the IP, the Center on Wrongful Convictions, and various scholars and researchers. The NRE also worked with the authors of the Encyclopedia of Wrongful Convictions and the creator of the website Justice Denied.
The major difference between the NRE and IP lists is that the NRE does not confine its cases to those of DNA exonerations but instead includes all exonerations. And that is why the NRE has become the “authoritative list” of all U.S. exonerations. The primary purpose of discussing the backgrounds of the IP and the NRE is that each group has its own standards for list creation and maintenance, and that led to coding discrepancies.
The NRE definition of exoneration pertaining to a formerly convicted person is intentionally conservative. The definition requires that a person be convicted; be relieved of all consequences of that conviction or any related convictions; that the relief be premised on new evidence of innocence; and there cannot be any “unexplained physical evidence of guilt.”
The focus of the interim report was the 342 listed by each organization as of 2017. It was not just within the 24 cases identified by LaPorte that coding discrepancies were found. The IP and the NRE identified five more exonerations that the IP had “coded with a forensic evidence contribution,” but the NRE hadn’t.
The IP and the NRE researchers also found 13 DNA exonerations in which the NRE reported forensic evidence contribution, but the IP hadn’t. In addition, both organizations consulted with the Convicting the Innocent’s DNA exoneration archive and found four additional cases that neither the IP nor the NRE had mentioned as having a forensic contribution used in the exoneration. This created a total of 46 cases for data coding to be reconciled.
The reconciliation process began on October 14, 2017, and by then, there were 351 total DNA exonerations recognized by both the IP and the NRE. An initial discrepancy that needed to be addressed for coding purposes was the agreement of the definitions before forensic science or forensic evidence could be coded as a contributing factor in the 46 cases being reconciled.
In 2017, the IP definition of “Misapplication of Forensic Science” (“MFS”) was to mean a case when it was known that “forensic evidence was used to associate, identify, or implicate someone who was later conclusively proven innocent with post-conviction DNA testing,” which would typically demonstrate the incorrectness of the original forensic evidence.
The NRE definition of “false or misleading Forensic Evidence” (“F/MFE”) meant an “exoneree’s conviction was based at least in part on forensic information that was (1) caused by errors in forensic testing, (2) based on unreliable or unproven methods, (3) expressed with exaggerated and misleading confidence, or (4) fraudulent.”
The two groups understood that first reconciling the definitions might not be easy or even possible. This process itself led to in-depth discussions and the creation of a “Forensic Advisory Group consisting of 14 experts” with divergent viewpoints on forensic evidence. These discussions led to the revision of the NRE’s F/MFE definition.
The revised F/MFE definition became “faulty of misleading expert or forensic evidence may have led to a factually erroneous conclusion, at any stage of the investigation or adjudication, that contributed to the false conviction.” The revised mutual definition that was created is identified under the heading “Forensic Contributing Factor” (“FCF”) as a blanket term referring to both MFSs and F/MFEs. In the end, all 46 cases were reconciled. Of those, 39 had an FCF, and the other seven did not.
The interim report also identified other difficulties in wrongful conviction research. There are substantial informational deficits. Although the two groups have access to large data pools, there is “incomplete information about every case.” And the NRE is a “living archive” that is constantly adding new cases. The NRE adds about one case per business day per year. New information about the 3,200 cases is added frequently as well. During the five-year reconciliation period, 24 new exonerations meeting the IP’s definition of DNA exoneration were added to that list alone. Six of these cases were found to have F/MFE as a contributing factor. The IP’s more specialized list had risen to 375 total DNA exonerations by April 28, 2022.
What the groups did find was that the coding changes trended toward identifying more problems with forensic evidence previously not known. The researchers believe that there is much more information about forensic contributions that wrongful conviction researchers simply aren’t aware of yet. Researchers discovered that rape or rape-murder convictions account for almost three-quarters of all DNA exonerations.
The awareness of flaws in forensic analysis is having a positive impact though. In the 1980s, there were 215 such wrongful convictions. But since 2000, there have been only 27 such convictions identified, and none of those has occurred in the past 12 years. DNA testing has “greatly increased” since 2000, and the proper use of this testing pre-trial, or oftentimes even pre-arrest, explains why “the share of exonerations that include DNA evidence has declined substantially.” The interim report describes this trend as the “singular accomplishment of forensic science.”
Beyond that accomplishment, very little of the interim report discusses the state of forensic science or how it can be improved for use in the legal system. What its authors do strive for is directing readers to information like LaPorte’s research article and the validity of his concluding statements. These statements provide common beliefs about forensic science and wrongful convictions.
These statements consist of the following beliefs: (1) forensic misconduct is wholly unacceptable, (2) the use of ambiguous terminology in reports and testimonies should be avoided by forensic scientists, (3) forensic scientists should always strive for objectivity and impartiality, (4) inevitable forensic errors should be occasions for learning, not diminished or hidden, (5) and the limitations of scientific analysis methods should be respected by forensic scientists. A report concerning the role of forensic evidence in all exonerations is being prepared by the groups.
As a digital subscriber to Criminal Legal News, you can access full text and downloads for this and other premium content.
Already a subscriber? Login