Since Bedau and Radelet’s study was published in 1987 and it received so much attention, including a critique directed at the 23 wrongful executions that they identifi ed, much of the research post-1987 has been dominated by exonerations of convictions that are less controversial and more objectively supported as “wrongful” by offi cial acknowledgment of a reversal or not guilty verdict by the crimi- nal justice system.
A fairly large number of articles, reports, and books, as well as documentaries, focusing on specifi c exonerations or on larger populations of cases have been produced. These studies tend to utilize a narrative methodology and are largely journalistic in orientation (Gardner, 1952; Yant, 1991; Frank & Frank, 1957). Much of this research has most commonly focused on single exonerated individuals. Examples include examinations of the cases of Randall Dale Adams (Adams, Hoffer, & Hoffer, 1991), Walter MacMillian (Earley, 1995), John Henry Knapp (Parloff, 1996), Clarence Lee Brandley (Davies, 1993), Anthony Porter (Armbrust, 2002), Joe Spaziano (Mello, 2001), Earl Washington (Edds, 2003), and the Ford Heights Four (Warden & Protess, 1998). As previously mentioned, the fi rst compilation of cases was conducted by Bedau and Radelet (1987); however, their pool of cases also included 23 wrongful executions. In an effort to be less controversial, in their updated book in 1992, they limited their cases to the exonerated. Moreover, in 1996, they presented brief vignettes of the 68 exoneration cases nationwide since 1972. The standard used by Radelet, Lofquist, and Bedau (1996, p. 914) to identify these cases involved “deferring to the decision made by the fi nal court or jury that evaluated the evidence in the case.” In their study, Radelet and his colleagues found that erroneously convicted defendants spent an average of approximately seven years on death row prior to release (Radelet, Lofquist, & Bedau (1996, 1996). Additionally, they noted that a majority of those released from death row were African Americans or other minorities. The Death Penalty Information Center (DPIC) has subsequently taken over the list of the exonerated and continually provides a short summary of each case on its website. At the end of 2016, the list contained 156 exonerations to date (DPIC, 12.12.16).
One of the fi rst studies to attempt to identify predictors of wrongful convictions was conducted by Harmon (2001a). Previous research on wrongful convictions in capital cases had focused primar- ily on qualitative methods designed to provide in-depth descriptive analyses of these cases (Borchard, 1932; Gardner, 1952; Radin, 1964; Bedau & Radelet, 1987). In contrast, Harmon (2001a) utilized a quantitative comparison between the 76 documented cases from 1970–1998 in which prisoners were released from death row because of “doubts about their guilt” (identifi ed by Radelet et al. ) and a matched group of inmates who were executed. Through the utilization of a logistic regression model, signifi cant predictors of cases that resulted in a release from death row as opposed to an execution were identifi ed. These signifi cant factors included the following factors: allegations of perjury of witnesses, the discovery of new evidence postconviction, type of attorney on appeal, and strength of the evidence.
Numerous scholars have continued to utilize a comparison group approach to study and advance knowledge in this area (see Leo & Gould, 2009, Gould & Leo, 2010). In their 2009 study, Leo and Gould argued that social scientifi c methods would allow for more accurate and exact depictions of the multifactorial and complex nature of causation in wrongful conviction cases. The authors discussed and illustrated several social science approaches to the study of wrongful conviction: aggregated case studies, matched comparison samples, and path analysis. They argue that utilizing all of these methods would assist criminal law and procedure scholars to better advance and understand the causes, char- acteristics, and consequences of wrongful convictions.
In their recent study of wrongful convictions that were not limited to capital cases, Gould, Car- rano, and Young (2013) utilized a comparison methodology to determine signifi cant factors that were present in cases that led the system to rightfully acquit or dismiss charges against innocent defendants (so-called “near misses”) that were not present in cases that led the system to erroneously convict the innocent. The results indicated that ten factors—the age and criminal history of the defendant,
32 Wrongful Capital Convictions