Show Summary Details
Page of

(p. 1) Introduction and Overview 

(p. 1) Introduction and Overview
(p. 1) Introduction and Overview

Matthew Barry Johnson

Page of

PRINTED FROM OXFORD CLINICAL PSYCHOLOGY ONLINE ( © Oxford University Press, 2021. All Rights Reserved. Under the terms of the licence agreement, an individual user may print out a PDF of a single chapter of a title in Oxford Clinical Psychology Online for personal use (for details see Privacy Policy and Legal Notice).

date: 25 January 2021

There can be no doubt that for every proven case of wrongful conviction there are many more that remain unproven.

Gisli Gudjonsson (2003, p. 163)

In 2010, Edward Carter was exonerated and released from prison after 35 years of confinement for a crime he did not commit. At age 19, he was convicted of the knifepoint rape of a white university student in Detroit, Michigan (Gross & Shaffer, 2012). The only evidence against Carter was the victim’s identification, though he had a verifiable alibi. After exhausting all appeals, Carter sought DNA testing through a Michigan Innocence Project. The crime scene DNA had been destroyed, but fortunately a police officer did further investigation, locating fingerprints from the crime scene. The prints were submitted to the FBI Automated Fingerprint Identification System and matched to a convicted sex offender then imprisoned for similar rapes at the university during the same time period. Carter reported that his trial public defender urged him to plead guilty and failed to effectively defend him when he refused. According to Carter, he was actually in police custody on an unrelated charge at the time of the offense; nevertheless, the victim picked him out of a photo spread that included multiple photos of him. The victim then identified Carter in a live lineup where the others did not fit the description of the perpetrator. Carter’s parents and girlfriend died while he was imprisoned. His major regret on release at age 55 was that he never had children. In 2017, he was awarded a 1.7-million-dollar settlement through recently enacted legislation (E. Anderson, 2017). It is worth noting that Carter was a suspect with prior criminal justice exposure and the offense he was convicted of was actually committed by a serial sex offender. Carter’s exoneration received virtually no media attention (Gross & Shaffer, 2012), but in many ways, his case is emblematic of wrongful conviction of sexual assault in the (p. 2) United States—an innocent Black defendant misidentified by a white victim of a “stranger rape.”

Wrongful Conviction and Rape

This book is focused on two sets of egregious violations, rape/sexual assault and wrongful criminal conviction. Both are tragic and traumatic occurrences with multiple adverse, chronic consequences. By far, the large majority of confirmed wrongful convictions in the United States have occurred in cases of rape and sexual assault where prosecutors were able to convict innocent suspects of major crimes, resulting in grave injustice for the defendants, the crime victims, and society at large. This book highlights characteristics of rape/sexual assault that increase the risk of wrongful conviction.


Rape and sexual assault produce severe acute and chronic injuries to victims. In some cases, victims are murdered (rape/murder) in the course of the crime.1 Estimating the extent of the damage from rape is difficult for a variety of reasons. Incidence rates assess the emergence of cases within a certain time frame, typically a year. Prevalence addresses the number of existing cases at a given time. Garland (2009), citing information from the Bureau of Justice Statistics for the mid-1990s, reported 335,000 US rape and sexual assaults a year, with only a third of victims filing reports with law enforcement. Garland also cited varying sources suggesting the lifetime prevalence of sexual victimization (a broader term) among women to be 12% to 50%. The use of multiple terms for these crimes (“rape,” “forcible rape,” “sexual violence,” “deviant sexual battery,” and so on) also contributes to a lack of reliable incidence and prevalence rates.

What is clear is that rape is a notoriously underreported crime. Typically, victims are unwilling to make official reports, and according to Garland, some (mistakenly) regard their assault as something other than rape. Ambiguity stemming from state variation in defining rape and prevailing cultural biases leads some victims to believe “real” rape is when she has “been brutally beaten by a minority stranger” (Garland, 2009, p. 8; also see Estrich, 1987). Rape victims are often confused, ashamed, fearful of the assailant, and also afraid that they will not be believed by friends, family, and law enforcement.

More recent Bureau of Justice Statistics data on rape and sexual victimization were provided by Planty, Langton, Krebs, Bersofsky, and Smiley-McDonald (2013). The encouraging news is that the rate of sexual violence decreased substantially (by 64%) from 1995 to 2005. The decrease was apparent across race/ethnic groups, though lower socioeconomic status continued to be associated with increased victimization. The large majority of victims (78%) knew their offender; stranger assaults occurred in only 22% of cases. Victims reported use of (p. 3) a weapon by the perpetrator in 11% of the cases. The percentage of victims who reported assaults to police increased from 29% in 1995 to 35% in 2010, with fear of reprisal as the most commonly cited reason for failure to report. The percentage of reports that produced arrests decreased from 47% to 31%.

During the decade of the 2000s, the US Centers for Disease Control and Prevention (CDC) promoted multifaceted initiatives to define rape and sexual violence as a public health issue and to support (primary) prevention research and intervention approaches (Degue et al., 2012). The indices cited previously of reduced sexual victimization suggest that the CDC’s initiatives, in conjunction with other efforts, contributed to a reduction in sexual violence and related harm; however, sexual violence remains a major source of adversity with traumatic and, at times, fatal outcomes.

Wrongful Conviction

While there is a lengthy history of research on wrongful conviction and miscarriage of justice (Bedau & Radelet, 1987; Borchard, 1932; Gould & Leo, 2010; Leo, 2005), the past 30 years have witnessed a growing academic, professional, and public interest associated with DNA science confirming wrongful convictions. Borchard’s 1932 book, Convicting the Innocent: Sixty-five Actual Errors of Criminal Justice, is regarded as the first academic study of wrongful conviction in the United States. The work documents and examines 65 wrongful criminal convictions and served as a rebuttal to influential circuit court Judge Learned Hand’s 1923 remark that, “The ghost of the innocent man convicted . . . [was] an unreal dream” (Krajicek, 2015).

As noted by Leo (2005), subsequent decades produced more wrongful conviction scholarship and journalist contributions. Typically, these works identified additional cases and echoed a familiar narrative that began with an innocent defendant tried and convicted despite the axiom that it is better to allow some number of criminals go unpunished than to convict one innocent. The narrative was completed by heroic efforts marshaled to exonerate and free the wrongly convicted. In 1987, Bedau and Radelet (1987) published a major law review that presented 350 wrongful convictions in capital cases dating back to the beginning of the 20th Century. While this work was very well regarded among scholars interested in wrongful conviction, the broader influence of the research was limited because it relied, in large part, on the authors’ judgments about innocence.

A good deal of informed literature has addressed the topic of how to identify and define wrongful conviction and exoneration (Bedau & Radelet, 1987; Gross, Jacoby, Matheson, Montgomery, & Patil, 2005; Gudjonsson, 2003). Identification of a wrongful conviction requires some type of process for evaluating and thus confirming the defendant was indeed innocent. Proving innocence in an absolute way may be difficult or impossible. Exoneration is an official process; one leader in the field, Gross (Gross, O’Brien, Hu, & Kennedy, 2014; Gross & Shaffer, 2012) used the term to refer to cases in which, following conviction, some kind (p. 4) of official statement is made that the defendant was not guilty. Examples are executive pardons based on evidence of innocence, charges dismissed based on evidence of innocence, and defendants acquitted at retrial where evidence of innocence was previously unavailable. Consistent with Gross, for the purpose of the current presentation, here the term wrongful conviction is used to refer to cases that have received an official declaration the defendant was not guilty. It is important to note there are several cases in which there is substantial evidence of innocence, but an official exoneration has not occurred for any number of reasons unrelated to criminal evidence.2

DNA Exonerations

The first DNA exoneration in the United States occurred in 1989. Gary Dotson was convicted in 1979 for the rape of Cathleen Crowell-Webb (Garrett, 2011; “Gary Dotson—Northwestern School of Law,” n.d.). The case was investigated as a stranger rape when actually there was no crime at all. Prior to the presentation of exculpatory DNA, the purported victim testified under oath in postconviction hearings that she lied about the accusation against Dotson; she fabricated the rape story to conceal her sexual activity with her boyfriend because she feared she was pregnant (she was 16 years old at the time). However, her testimony was not sufficient for the exoneration. Mr. Dotson’s release, after 10 years, occurred when DNA evidence was presented at subsequent hearings. In addition to the intentionally false rape charge, the case involved faulty scientific testimony about serology (blood typing) and hair analysis, as well as apparent prosecutorial misconduct.3 It should be noted that “false rape charges” as in the Dotson case are not typical of wrongful convictions in sexual assault.

The application of DNA science to criminal investigation has altered the meaning and perception of criminal evidence in many ways. First, DNA science has made it much more difficult to ignore or dismiss the reality of wrongful criminal convictions in the US legal system. As DNA analysis has become increasingly more efficient (Scheck, Neufeld, & Dwyer, 2001), it has provided conclusive evidence of guilt in many cases, as well as innocence in others. But DNA analysis can only be applied to the small proportion of criminal cases (5%–10%; see Innocence Project, 2009) in which human bodily residues (blood, sweat, tears, semen, saliva, hair, etc.) are taken into evidence. Nevertheless, the yield from DNA analysis extends beyond the particular cases in which it is applied. Postconviction exculpatory DNA evidence has demonstrated that major forms of criminal evidence typically relied on to prove guilt (e.g., eyewitness testimony, informant testimony, confessions, hair, fiber, footprint, and dental bite mark impression analyses) are often unreliable (Garrett & Neufeld, 2009). This has profound relevance to virtually all criminal prosecutions regardless of the existence of DNA evidence in any given case.

Manufactured evidence is a term that has been a staple of wrongful conviction scholars and researchers as far back as Borchard (1932). While the prevailing (p. 5) perception of criminal investigation is that evidence is discovered and collected, the reality is that evidence presented to prove guilt is often created or “manufactured” by the police or prosecution. The manufacturing of evidence occurs via concealed interrogation and witness identification procedures, secret (or otherwise unacknowledged) incentives provided to informants for testimony, undisclosed informal relationships among prosecutors and forensic experts, and other means. The failure to objectively preserve the evidence from these procedures amounts to a suppression of evidence. The term black box is used in this book to characterize the secret domains where evidence is manufactured. As noted, the primary two black boxes are unrecorded custodial interrogations and unrecorded witness identification procedures, though there are others.

A major milestone in wrongful conviction research and awareness occurred with the 1996 publication Convicted by Juries, Exonerated by Science: Case Studies in the Use of DNA Evidence to Establish Innocence After Trial (Connors, Lundregan, Miller, & McEwen, 1996). This report, commissioned by the National Institute of Justice, presented findings from early cases in which postconviction DNA evidence was used to exonerate defendants convicted of major offenses. Attorney General Janet Reno noted in the Preface, “The criminal justice system is not infallible, and this report documents cases where the search for truth took a tortuous path” (p. iii). The authors’ comment, “All 28 cases profiled in the report involved some form of sexual assault” (p.12) is remarkable for what is stated, as well as what was unstated. All the cases did include “some form of sexual assault” allegation; however, in at least one case (Gary Dotson’s) there was no rape at all. In two others (Mark Bravo and Gerald Davis) it was not clear whether a rape occurred. Bravo was convicted in 1990 of raping a patient in a psychiatric hospital where he worked. Postconviction DNA testing matched none of the crime scene semen to Bravo; furthermore, the complainant recanted her testimony, so it was not apparent whether this was a misidentification or a no-crime case (Connors et al., 1996). Davis was convicted of raping a woman who reported that she was a family friend. Trial testimony from discredited forensic chemist Fred Zain4 indicated Davis could not be eliminated as a source of semen recovered from the complainant, but postconviction DNA results excluded Davis as a contributor of the crime scene semen. Thus, again, it was not apparent if this was a no-crime case or misidentification. A more substantial finding in the report, though not articulated by the authors, was that 25 of the 28 cases were stranger rapes as opposed to the more common “acquaintance” or “date” rape (or they were mistakenly prosecuted as stranger rapes, as in the Dotson case).

As noted, available data strongly suggest that the less frequently occurring type of rape, stranger rape, is disproportionately associated with wrongful conviction when compared to the more common “acquaintance rape.”5 By and large, informed commentators, such as Neufeld and Scheck (as cited in Connors et al., 1996), viewed the matter of wrongful convictions being concentrated among “some form of sexual assault” to be an artifact of the availability of DNA evidence in sexual assaults. This left unexamined and underappreciated the unique features of stranger rapes (with regard to the offenses themselves, investigation into the (p. 6) crimes, and prosecution) that increase the risk of wrongful conviction. These features include, but are not limited to, pressure on law enforcement officials and the enhanced risk of misidentification. Multiple sources indicated common misidentification error to be in the range of 30% (Gaulkin, 2010), and this is likely increased where there is traumatic victimization (Bookbinder & Brainerd, 2016; Deffenbacher, Bornstein, Penrod, & McGorty, 2004). Further, additional obstacles to reliable identification exist when the witness and subject are from different race/ethnic groups—cross-racial identification (Connelly, 2015)—as elaborated in Chapter 3.

Several researchers (Free & Ruesink, 2012; Gross et al., 2005; Gross & Shaffer, 2012) have noted that the most common offense associated with wrongful conviction is murder/homicide, and the second most common offense is rape/sexual assault. But, this is an artifact of classifying cases by the most serious charge, the classification approach employed for purposes of criminal prosecution. The focus of this book is on classification that informs processes involved in wrongful conviction.

As presented in Figure I.1, when rape/sexual assault cases are combined with murder/homicide cases that included sexual assault charges, it becomes evident that the majority of confirmed wrongful convictions in the United States have involved rape/sexual assault charges (89% of the DNA exonerations; Garrett, 2011). This is especially significant given that sexual offenses account for only about 10% of the prison population (Gross et al., 2005).

Figure I.1 Percentage of Offenses Among the First 250 DNA Exonerations (from Garrett, 2011).

Figure I.1 Percentage of Offenses Among the First 250 DNA Exonerations (from Garrett, 2011).

The Frequency of Wrongful Conviction

How common is wrongful conviction? Virtually all researchers (Garrett, 2011; Gross et al., 2014; Gross & Shaffer, 2012; Kassin et al., 2010; Leo, 2008; Neufield & (p. 7) Scheck as cited in Connors et al., 1996) agreed that known or confirmed wrongful convictions represent only a fraction of innocent defendants found guilty at trial or by guilty plea. Several investigators have sought to estimate the frequency of wrongful convictions. Krajicek (2015) referred to calculation of the aggregate number of wrongful convictions as “the Holy Grail” question in the field. Huff, Rattner, and Sagarin (1996) offered an estimate of 0.05%–1.0%, drawing on a survey of judges, attorneys, and law enforcement officials in Ohio. While this suggests a high degree of accuracy, it also would mean at least 10,000 wrongful convictions among the 2 million defendants convicted a year at the time of their study. Risinger (2007), critiquing Supreme Court Justice Scalia’s suggestion of a 0.027% rate of wrongful criminal convictions in the United States, calculated “a minimum factually wrongful conviction rate” (p. 762) for rape-murders as 3.3% and suggested the rate across all crime could be as high as 5%.

Gross and O’Brien (2008) found a 2.3% exoneration rate among defendants sentenced to death from 1973 to 2003 but noted the rate could not be generalized beyond their particular sample. Gross and O’Brien described enormous methodological obstacles in determining the prevalence of false convictions and stated, “It will be difficult to learn more” (p. 958).6 Leo and Gould (2009), critical of the pessimism conveyed in the Gross and O’Brien formulation, pointed out that the unknown aggregate wrongful conviction figure need not deter ongoing empirical research to discover, understand, and prevent wrongful conviction, noting other criminological research advances in the context of unknowns. In addition, Leo and Gould recommended the adoption of social science methods to capture the multifactorial and contingent interaction of variables that contribute to wrongful convictions. In a subsequent publication, Gross et al. (2014) reported a new analysis of US death sentences from 1973 to 2004 and found that 117 (1.6%) of the defendants had been exonerated. Gross and his colleagues estimated that another 200 prisoners from this sample may have been innocent but were without access to the legal resources to demonstrate their innocence. According to these researchers, this was suggestive of a 4.1% wrongful conviction rate, though Gross and his colleagues referred to the aggregate wrongful conviction rate as “unknowable” (Gross et al., 2014, p. 7230).

This limited review indicates estimated rates of wrongful conviction from leading investigators ranging from 0.05% to 5%. As Leo and Gould (2009) suggested, arguably the aggregate wrongful conviction rate is not necessary to further advance meaningful research. However, it is an important metric if only because critics (e.g., US Supreme Court Justice Scalia, as noted previously) attempt to dismiss or marginalize the investigation of wrongful criminal conviction in the United States, characterizing it as a rare aberration, and promote an image of the US legal system as fair, impartial, and efficient. Stated differently, these critics aim to advance the ideal of the US legal system as “imperfect but the best in the world” and thus to curtail what are characterized as frivolous postconviction appeals, providing closure and finality to legal rulings. Thus, in many ways, wrongful conviction is an area of investigation that runs contrary to deeply held beliefs about US government institutions (police and law enforcement, “first responders,” and (p. 8) the court system) and the narrative of a legal system based on democratic ideas of liberty, justice, and equal protection of law.

Disaggregating Wrongful Conviction

The current project is guided by the Leo and Gould (2009) formulation that despite obstacles, much has been learned and much more can be discovered about wrongful conviction. The current examination asserts the increasingly visible wrongful convictions in sexual assault (a major portion of all wrongful convictions) can be reduced substantially through recognition of the unique wrongful conviction risks in different types of rape/sexual assault cases, that is, “stranger rape” versus “date/acquaintance rape” versus “child sexual abuse” cases. While stranger rapes garner the most press and public attention (and fears), they are much less common than date and acquaintance rapes (Maston & Klaus, 2003; Planty et al., 2013). The confirmed wrongful convictions in sexual assault are highly concentrated among stranger rapes and associated with faulty identifications and police-induced false confessions (as well as other factors). Wrongful conviction in date and acquaintance rape is associated with false rape charges, is less common,7 and is associated with complainant characteristics (which converge with other factors). Child sexual abuse cases can also be divided in stranger versus acquaintance assaults, but in addition there are special risks associated with the manner in which the victim/complainants are assessed and interviewed, interpersonal discord associated with divorce and separation, plea bargaining in the context of severe penalties, and social hysteria associated with many prosecutions. These features are presented in greater detail in subsequent chapters.

An overview of the general disaggregation approach used in this study is illustrated in Figure I.2. There are confirmed/exonerated wrongful convictions and an unknown amount of unconfirmed wrongful convictions. Among the confirmed wrongful convictions, there are sexual assaults (including sexual assaults that include murder), murders, and other offenses. Among the sexual assaults, there are “stranger rapes,” “acquaintance rapes,” “intrafamilial child sexual assaults,”8 and cases where the offense type was undetermined. Among the stranger rapes, there are differential risks associated with cases with capable victims versus cases with incapacitated victims.

Figure I.2 Disaggregating Wrongful Conviction Diagram.

Figure I.2 Disaggregating Wrongful Conviction Diagram.

Diagram prepared by Riki Martinez

So, through disaggregating the wrongful conviction database by offense, and grouping the sexual assaults together (rather than categorizing the rape/murders as murders), it is apparent a distinctively large proportion of cases involve rape/sexual assault. Further disaggregation indicates the wrongful convictions are not distributed randomly among rape and sexual assaults but occur disproportionately among the less common type of rapes, stranger rape. In disaggregating by offense, we are not so much concerned with the total aggregate number but rather with the distribution among a population. Disaggregation enlists a public health or epidemiological perspective. For instance, in examining suicide, we want to know not only the aggregate suicide rate but also the rate is higher among (p. 9) adolescent males, and maybe higher still among Latina adolescents, or among the elderly with certain chronic conditions. In the examination of automobile or airline accidents, it is valuable to know the general rate of occurrence and even more valuable to know the general rate is increased in the presence of certain risk factors (weather, driver/pilot fatigue and distraction, substance abuse, the design of roadways or airports, and so on).

Disaggregating the wrongful conviction database by race/ethnicity reveals other important findings. For instance, a disproportionate concentration of Black and Latino defendants are wrongfully convicted. Garrett (2011) reported that 75% of rape exonerees are Black or Latino, while these ethnic groups make up only 30%–40% of rape convicts. Further disaggregation based on race/ethnicity of both victim and defendant indicates a concentration of wrongful convictions in sexual assaults associated with white victims of stranger assaults by Black and Latino assailants, a decidedly rare offense. Alternatively, if we disaggregate by the circumstances of conviction, we see that a number of innocent defendants pled guilty to rape (M. B. Johnson & Cunningham, 2015), which raises questions about the operation of the judicial system. Examining wrongful conviction in stranger rape, we find that a considerable portion of the offenses are actually committed by serial rapists, which warrants a focus on characteristics of the offense and investigation that contribute to the prosecution of innocent suspects.

(p. 10) Gross et al. (2005) reported disaggregation of wrongful conviction by state and found wrongful convictions appear to be concentrated in certain states (Gross et al., 2005). Further research (Gross & Shaffer, 2012) revealed that state variations in wrongful conviction rates masked more substantial variability at the county level, and this invites questions about organizational culture and practices in certain district attorney/prosecutor’s offices (R. King, 2016). While the risk of wrongful conviction exists, there are coexisting risks of the perpetrator escaping prosecution. Disaggregation of the existing wrongful conviction data suggests these risks are differentially distributed between two major categories of rape/sexual assault. The risk of wrongful conviction is higher among stranger rapes, and the risk of the perpetrator escaping conviction is higher among date and acquaintance rapes. However, generally wrongful conviction in sexual assault means the actual offenders remains free, and, as discussed in subsequent chapters, these offenders commonly assault other victims.

Why Study Wrongful Conviction in Sexual Assault?

The purpose of Wrongful Conviction in Sexual Assault is to contribute to the growing body of knowledge that can be used to prevent wrongful convictions through pretrial and trial intervention; to correct wrongful conviction through postconviction relief appeals; and to generate policy recommendations that will reduce the occurrence of wrongful convictions. In addition, the findings presented here will be useful in clarifying potential flaws in the US criminal justice system and suggesting alternative approaches and models of adjudication and intervention.

Outline of the Chapters

This book proceeds by examining wrongful conviction in sexual assault from several perspectives. Chapter 1 provides an overview of the rape law reform movement that began in the 1970s and continued at least through the 1990s. The movement sought a series of reforms designed to encourage victims to make official reports that would facilitate the successful prosecution of rapists and sex offenders. The movement was quite effective in achieving legislative reform, but there was no discernible increase in sexual assault convictions. The rape law reform movement solidified an alliance of concern that strengthened vigorous prosecution of stranger rapes but had little impact on the more common type of rape, acquaintance rape.

Chapter 2 focuses on the concentration of rape cases among confirmed wrongful convictions, examining how stranger rape differs from date and acquaintance rape with regard to risk of wrongful conviction. Chapter 2 also examines the pressures on law enforcement authorities and the roles of primary evidence (misidentifications and false confessions); and secondary evidence, “black box” (p. 11) investigation methods and “victim status” in stranger rape. In addition, Chapter 2 presents a “stranger rape thesis” to distinguish the unique challenges faced in the investigation of stranger rape. The moral outrage that tends to result from stranger rape results in great pressure on police for arrests and convictions, yet reliable identification of the perpetrator is often compromised in stranger rape.

Chapter 3 examines the current disproportion of Black defendants wrongly convicted of sexual assault through a historical lens, noting the history in the United States of statutorily separate sexual assault penalties based on race. Chapter 3 presents how race, rape law, and prosecution have been manifest in different historical eras in the United States and how racial bias against Black defendants continues in the post–civil rights era.

Chapter 4 describes a common pattern in which innocent defendants are tried and convicted of crimes committed by serial rapists. This chapter highlights the difficulty encountered by law enforcement in the investigation of stranger rapes despite the expanding literature on crime scene investigation and offender profiling.

Chapter 5 reviews the series of day care–based child sexual abuse prosecutions that occurred in the United Sates from the mid-1980s to the mid-1990s. Charges in these cases often relied on highly implausible accusations. Suggestive questioning of children, prosecutorial overcharging, and a prevailing social hysteria contributed to a substantial number of innocent defendants being convicted.

Chapter 6 discusses concepts and models used to describe and explain processes in wrongful criminal conviction, such as “tunnel vision,” “confirmation biases,” “misinformation effects,” and “escalation of commitment.” This chapter considers the research focus on “causes” or contributors to wrongful conviction with an emphasis on trial evidence and variables; it also draws on several theoretical sources to introduce the notion of wrongful conviction as a form of “moral correction” that results in emotional and cognitive factors that adversely affect the evaluation of evidence and contribute to the risk of wrongful conviction.

Chapter 7 summarizes the findings from the previous chapters and examines recommendations for reform from across the field. Chapter 7 not only makes note of the obstacles to reducing wrongful conviction but also highlights the guidance that has emerged from the expanding academic research and legal engagement in the past two decades.

The wrongful conviction data reported and relied on in this book are provided online ( This includes the following appendices:

  1. A Wrongful convictions in sexual assault as reported on the Innocence Project website

  2. B Wrongful convictions in sexual assault as reported on the National Registry of Exonerations

  3. C Exonerees wrongly convicted of sexual assaults committed by serial sex offenders

  4. D Wrongful convictions related to the child sexual abuse hysteria as reported on the National Registry of Exonerations


1. Michelle Bosko (Chapter 4), Lori Roscetti (Chapter 3), Nancy DePriest (Chapter 2), Angela Corea (Chapter 2), Deborah Sykes (Chapter 3), Tyrone and Tina Urquhart (Chapter 3) were all rape/murder victims, and innocent defendants were convicted of the crimes.

2. See the discussion by Gross et al. (2005) of “foot dragging” by government officials, guilty pleas with release provisions offered to avoid risk of retrial, and other situations amounting to negligence. Also see where exonerations are bargained for terms acceptable to the original prosecuting authorities, as in the Kerry Max Cook case (M. Hall, 2017).

3. In closing arguments, the prosecutor stated the forensic analyst “matched” Dotson’s hairs to the victim when the analyst only said they were similar (Garrett, 2011).

4. As noted in Connors et al., 1996, Fred Zain provided unsubstantiated incriminating testimony for the prosecution in several cases. Also see the work of Garrett and Neufeld (2009).

5. Another observation made by Connors et al. (1996) was that a substantial proportion of the defendants (15 of the 28) were known to police prior to arrest, suggesting prior police contact may be associated with vulnerability to wrongful conviction.

6. However, Gross and O’Brien reported a “strong demographic pattern” where “black men accused of raping white women face a greater risk of false conviction than other rape defendants” (Gross & O’Brien, 2008, p. 958).

7. False rape charges are rare though highly publicized when identified, as in the Gary Dotson case as well as the Brian Banks case presented in Chapter 2.

8. In addition, there are intrafamilial adult sexual assaults that are no doubt underreported and are not addressed in this work.