Abstract
To improve criminal justice responses to sexual assault, it is vital to understand the factors that influence decisions made by police and prosecutors and how these decisions overlap. Although decisions made at the prosecutorial stage are largely reliant on the actions of the police, the prosecutors’ charging decisions often differ from how the police classify incidents. Guided by the focal concerns perspective, this study examined 231 cases of adult sexual assault incidents reported to the police to identify which factors were predictive of congruent charging decisions and arrest. The results supported the focal concerns perspective and revealed that legal and extralegal factors were influential at both of these points. Legally relevant factors that focused on the collection of evidence were paramount in decisions at both stages; however, extralegal variables such as timely reporting and whether the victim physically resisted were also significantly related to decisions at each stage. Implications arising from the results are discussed.
Introduction
Research has illuminated that a very small proportion of sexual assault victims report the incident to police (Tjaden & Thoennes, 2006). When victims do report, case attrition remains high at both the police and prosecutorial decision-making points (Alderden & Ullman, 2012). The predominance of rape myths and victim-blaming views in American culture may partially explain why reporting rates are low and attrition rates are high when compared with other crimes. The handling of cases once they enter and move through the criminal justice system may also be responsible for the high dropout rate and the reluctance to report sexual assault victimizations.
Prior research has demonstrated that police and prosecutors both exercise considerable discretion when deciding how to pursue sexual assault cases (Frazier & Haney, 1996; Frohmann, 1991; Holleran et al., 2010; Kelley & Campbell, 2013; Kerstetter, 1990). The decisions made by police officers dictate the workload for prosecutors as the prosecutors are charged with reviewing the cases that are forwarded to them by the police. In turn, prosecutors use their discretion when determining which cases to accept or reject and, if accepted, whether to agree with the police classification of the crime or file different charges. Ultimately, the police and prosecutors work toward the same goal: to move cases that are perceived as winnable through the criminal justice system. Although the police and prosecutors are interconnected in their pursuit of winnable cases, only a small body of research exists to inform our understanding of the decision-making dynamics between the two (Alderden & Ullman, 2012; Holleran et al., 2010; Kerstetter, 1990; Pattavina et al., 2016; Spohn & Tellis, 2018). To gain clarity about sexual assault case attrition and to inform policy, it is crucial to examine how these two branches influence each other’s decisions. While the police are traditionally viewed as the “gatekeepers” of the criminal justice system, research must consider how decisions made by prosecutors may also influence the way sexual assault cases are handled by the police.
Using police reports and court records, the current study examines the case attributes that predict the congruence between the police classification of crimes and prosecutorial charging decisions in adult sexual assault cases with victims who are willing to pursue criminal justice action. This study also examines whether the same case attributes are predictive of arrest. Guided by the focal concerns perspective and drawing from previous research, including Susan Estrich’s (1987) conception of “real rapes” and “genuine victims,” the current study contributes to the existing body of research by examining the decisions made by both the police and prosecutors in sexual assault cases and determining the extent to which the focal concerns perspective can apply to these key points. The focal concerns perspective was originally developed to describe how judges’ sentencing decisions were influenced by three focal concerns: their perceptions of offender culpability, the desire to protect the community, and practical constraints, or social costs of sentencing decisions (Steffensmeier et al., 1998). This perspective has since been extended to explain decisions made by prosecutors and police; however, with few exceptions, studies examining prosecutor and police decisions on the same cases are lacking. To date, the current study is the third study to apply the focal concerns perspective to the overlapping decisions of the police and prosecutors in all sexual assault cases that are referred to prosecutors, including cases that were referred prior to the arrest of an identified suspect. To date, this is the first study to examine the decisions of police and prosecutors without a specialized sexual assault unit in a medium-sized city. Although two prior studies have made valuable contributions to our understanding of police–prosecutor decisions in large jurisdictions, it is important to consider that 97.4% of local police departments in the United States served populations of less than 100,000 people in 2013 (Reaves, 2015). Similarly, many police departments, especially those in small and medium-sized jurisdictions, lack the necessary resources to develop and maintain specialized units to address sexual assault complaints. The current study advances what is known about charging and arrest in medium-sized jurisdictions by examining police and prosecutors’ decisions in sexual assault cases involving victims who were willing to pursue legal action.
The following sections include a brief review of prior research on sexual assault case processing. Police and prosecutorial decision-making in sexual assault cases are discussed, followed by a review of the focal concerns perspective and how it applies to police and prosecutor decision-making in sexual assault cases.
Review of the Literature
Sexual assault is a difficult and complex crime for the criminal justice system to process as many sexual assaults do not contain clear evidence that the crime occurred. Estrich (1987) described the concept of a “real rape” as a stereotypical attack in which a stranger appears from a dark alley with a weapon and forces sex onto a victim who suffers physical injuries from the attack. In a “real rape,” there is little doubt that the incident took place as the unsuspecting victim is a morally respectable female who was left with physical injuries from a violent attack. “Real rapes,” however, are the least likely to occur (Rape, Abuse, and Incest National Network [RAINN], 2013). More common are sexual assaults committed by intimates or acquaintances in the home of the victim or suspect (Jordan, 2004; RAINN, 2013). In addition, many sexual assaults lack physical evidence to indicate the incident took place (Jordan, 2004). As a result, the police and prosecutors rely heavily on victim accounts of the incident and their own perceptions of rape and who constitutes a rape victim when processing sexual assault cases (Galvin & Polk, 1983; Wentz & Archbold, 2012).
Police Officers’ Processing of Sexual Assault Cases
As the gatekeepers to the criminal justice system, police officers are granted a great deal of discretion in how to investigate and process sexual assault complaints (Kerstetter, 1990). As the initial point of contact in the criminal justice system for most sexual assault victims, police officers can choose whether to document a report, how much time and effort to devote to a case, whether to make an arrest, and how to classify a case when sending it to prosecutors. Ideally, police decisions are based solely on legally relevant factors, such as the seriousness of the offense (i.e., weapon use and victim injuries) and the amount of evidence supporting that a sexual assault occurred (e.g., forensic evidence, witness corroboration, and rape kit). Not surprisingly, early and recent studies have confirmed that identified suspects were more likely to be arrested in sexual assaults that involved a weapon (Bouffard, 2000; Kerstetter, 1990; LaFree, 1989; Spohn & Tellis, 2018), collateral injuries to the victim (Frazier & Haney, 1996; Rose & Randall, 1982), and forensic evidence that was collected, including evidence collected from a rape kit (Bouffard, 2000; Tasca et al., 2013).
Along with evidentiary factors and crime seriousness, research has revealed a multitude of extralegal variables that have affected the way sexual assault cases are handled by the police. Extralegal factors include characteristics of the victim, suspect, and incident that are beyond the legal scope of the offense. In other words, extralegal factors should not matter when determining whether a suspect committed the crime. In sexual assault cases, most of the extralegal factors revolve around the victim’s credibility (Alderden & Ullman, 2012; Beichner & Spohn, 2012; Frazier & Haney, 1996; Jordan, 2001, 2004; Page, 2008; Schuller & Stewart, 2000). Examples of extralegal factors found to influence police decisions in sexual assault cases include the relationship between the victim and suspect, the victim’s activity/behavior prior to the incident, the time it takes the victim to report the incident, and whether the victim physically resisted the incident (Bouffard, 2000; Frazier & Haney, 1996; Horney & Spohn, 1996; Kerstetter, 1990).
Research has shown the relationship between the victim and offender to be an important predictor of police decisions in sexual assault cases. Sexual assaults involving strangers fit the concept of an “ideal rape,” whereas the credibility and truthfulness of the victim are scrutinized if the victim is an acquaintance or intimate partner (Bryden & Lengnick, 1997; Estrich, 1987; Frazier & Haney, 1996; Spohn & Tellis, 2018). Not all studies have come to the same conclusion regarding stranger/nonstranger sexual assaults. Some research has found that offenders who were acquaintances of the victim were more likely to be arrested as apprehension is easier when the offender has been identified (Bouffard, 2000; LaFree, 1989). Police are also less suspicious of victims who report the assault immediately after it occurs (Jordan, 2001; Rose & Randall, 1982). In an examination of 610 sexual assault police incident reports, Rose and Randall (1982) found that the police interpreted victim hesitation to report in one of two ways: uncertainty about whether they were even assaulted, or that the victim needed time to create a “cover story” to hide their behavior (as in the case of infidelity). In addition, delayed reporting also reduces the ability to collect physical or forensic evidence.
Another crucial factor in police decision-making is whether the victim was engaged in risky behavior prior to the incident. Victims who used alcohol prior to the commission of the assault are perceived by the police as being partially responsible for the assault as drinking alcohol reflects poor judgment (Fisher et al., 2003; Holmstrom & Burgess, 1978; Jordan, 2004; Schuller & Stewart, 2000; Tasca et al., 2013). Subsequently, such cases are viewed as “weak” and receive less investigative effort (Holmstrom & Burgess, 1978). Overall, research shows that police officers consider a myriad of legal and extralegal factors when deciding how to respond to sexual assault cases. They are also more likely to pursue cases that they believe will be accepted by prosecutors because they are perceived to be “winnable” (Martin & Powell, 1994).
Prosecutors’ Processing of Sexual Assault Cases
Just as the police are considered the “gatekeepers of the criminal justice system,” prosecutors “hold the keys to the courthouse” (Neubauer, 1988, p. 200). Prosecutors use the information about cases that they receive from the police to determine whether to prosecute and which, if any, charges to file (Albonetti, 1986). Their reasons for not prosecuting are rarely ever reviewed, and cases that are not prosecuted end their path within the criminal justice system at that point (Spohn & Holleran, 2001). Prosecutors’ concerns with convictability in sexual assault cases create what Frohmann (1997) termed a “downstream orientation” by which prosecutors incorporate legal and extralegal factors in their decisions in an attempt to predict how cases will be perceived by judges and juries.
Sexual assault cases accepted by prosecutors are guided by legal and extralegal factors similar to the variables that have been noted in police research, and all serving as ways for prosecutors to assess the level of convictability for each case. With regard to legally relevant factors, research has shown that prosecutors were more likely to file charges in sexual assault cases that contained substantial physical or forensic evidence (Frazier & Haney, 1996; Kerstetter, 1990), cases in which weapons were used (Kerstetter, 1990; LaFree, 1981; Spohn & Spears, 1996), cases with witnesses present (Beichner & Spohn, 2012), and cases in which the victims sustained physical injuries (Alderden & Ullman, 2012; Burt, 1980; Frazier & Haney, 1996; McGregor et al., 2002).
Although prosecutors’ case decisions are guided primarily by legally relevant factors, such as the amount of evidence and the seriousness of the offense (Holleran et al., 2010), extralegal factors that pertain to the victim’s character and credibility are also considered, such as the victim’s relationship with the defendant (Beichner & Spohn, 2012). The impact of the relationship between sexual assault victims and suspects in prosecutors’ charging decisions remains uncertain. Some studies have found that sexual assaults occurring between strangers were investigated more thoroughly and are more likely to be charged as they tend to fit the description of what Estrich terms “real rapes” (Albonetti, 1986; Kerstetter, 1990; Weninger, 1978). Conversely, other research has concluded that the relationship does not influence prosecutors’ charging decisions (Horney & Spohn, 1996; Kingsnorth et al., 1999; Spohn & Spears, 1996). Kingsnorth et al. (1999) studied 467 sexual assault cases from prosecutorial intake through the sentencing disposition stages and found that the relationship between the victim and defendant did not affect the prosecutors’ decisions to accept the case.
Other extralegal factors linked to prosecutorial charging include whether the victim physically resisted the attacker and the risk-taking behavior of the victim. Charges are less likely to be filed in cases where the victim did not actively resist the attacker (Spohn et al., 2001; Spohn & Horney, 1993; Spohn & Spears, 1996) and in cases in which the victim engaged in risky behavior such as alcohol use prior to the incident (Beichner & Spohn, 2012; Spohn & Spears, 1996). Estrich (1987) asserted that blame and believability of the victim are greater factors in simple sexual assault cases as they are not considered “real rapes.” A simple rape usually occurs between acquaintances or intimate partners, rarely involves a weapon, and the victim rarely sustains injuries. Conversely, a real rape, or aggravated rape, is often more violent in nature, occurs among strangers, and often results in physical injuries (Spohn & Holleran, 2001). Victims’ morals and behaviors are more likely to be questioned in simple sexual assaults, as Spohn and Holleran (2001) illustrated in the following example: if a woman was raped by a complete stranger who held a gun to her head or a knife to her throat, criminal justice decision makers normally would not care whether she had been using drugs or was employed as a topless dancer. (p. 656)
As the majority of sexual assaults occur between acquaintances, it is likely that prosecutors use factors that question the victim’s character alongside legal factors when determining whether to accept or reject sexual assault cases.
The Focal Concerns Perspective
Originally used to describe disparities in judicial sentencing decisions (Steffensmeier et al., 1998), the focal concerns perspective may also be relevant to police and prosecutorial decision-making. The focal concerns perspective posits that the sentencing decisions of judges are based on three focal concerns that incorporate both legal and extralegal factors: (a) the blameworthiness of the offender, (b) the desire to protect the community, and (c) concerns about the social costs, or practical constraints and consequences, of their sentencing decisions (Steffensmeier et al., 1998). Blameworthiness is associated with the culpability of the defendant, the desire to protect the community emphasizes the dangerousness of the offender, and practical constraints focuses on the social costs and costs to the system that are associated with imprisonment. As judges rarely have enough information to accurately determine how dangerous or blameworthy the defendants are, they develop a “perceptual shorthand” that incorporates a number of extralegal variables that focus on the characteristics of the offender (e.g., age, race, sex) into their decision-making processes (Albonetti, 1991; Hawkins, 1981; Steffensmeier et al., 1998). This means that because of limited time and resources, and the lack of information about legally relevant qualities of a case, judges’ decisions include stereotypes and past decisions (Albonetti, 1991).
Holleran et al. (2010) extended the focal concerns perspective to the decision-making processes of the police and prosecutors in sexual assault cases. They argued that the focal concerns of the police and prosecutors are similar to those of judges; both are concerned with offense seriousness, the level of harm caused to the victim, and the blameworthiness of the offender. Holleran et al. (2010) and Spohn and Tellis (2018) noted one primary distinction between the practical constraints and consequences of judges and those of police and prosecutors. Whereas judges are concerned with the social costs of incarceration, police and prosecutors are concerned with the likelihood that the case will result in a conviction. Consequently, the perceptual shorthand that police and prosecutors develop in sexual assault cases involves stereotypical factors about what constitutes a “real rape” and a “genuine victim” as these are more likely to lead to convictions in cases that are taken to trial. Therefore, in addition to offender and incident characteristics, this perceptual shorthand also focuses on victim credibility factors, such as the behavior and background of the victims and the relationship between the victim and the suspect (Holleran et al., 2010). In this regard, when a victim does not conform to the “genuine victim” stereotype, the victim’s credibility is negatively affected, and the offender is viewed as less culpable for the crime.
Numerous studies have examined police and prosecutorial decisions in sexual assault cases separately; however, very few have tested the level of charging agreement between these two parts of the criminal justice system. Holleran et al. (2010) applied the focal concerns perspective in their analysis of two jurisdictions that each had specialized units devoted to sexual assault, but different sexual assault case screening criteria: Kansas City, which used a trial sufficiency standard (i.e., cases were assessed on the likelihood of conviction at trial), and Philadelphia, which used a legal sufficiency standard (i.e., cases were assessed based on the statutory elements of the crime). The authors found that charging agreement between police and prosecutors in both jurisdictions was guided primarily by the strength of evidence in the cases; however, extralegal factors were also considered. In Kansas City, but not Philadelphia, risk-taking behaviors of the victim reduced the likelihood that prosecutors would agree with the charging decisions of the police. The authors noted that police officers in Philadelphia worked very closely with attorneys and rarely were cases rejected once they reached the prosecutorial stage. More recently, Spohn and Tellis (2018) found support for the focal concerns perspective in their analysis of sexual assaults reported to the Los Angeles Police Department and the Los Angeles Sheriff’s Department. Their study showed profound interplay between police and prosecutors as the police often referred cases they believed prosecutors would reject for a prearrest charge evaluation prior to arresting suspects. Common characteristics of cases referred for a prearrest charge evaluation included the following: risky behavior by the victim, the victim did not report the incident promptly, the victim withdrew cooperation, the suspect did not use a weapon, physical evidence was absent, and no witnesses were present to corroborate the allegation (Spohn & Tellis, 2018). The authors noted that most of the cases referred for a prearrest charge evaluation were rejected for prosecution, which illustrates the importance of studying the extent to which the police and prosecutors work in tandem to process sexual assault cases.
Current Study
The findings from previous research demonstrate the importance of understanding the nexus between the decisions made by both police and prosecutors. Prior studies on charging agreement in sexual assault cases have focused on large cities and police agencies that have specialized sexual assault units or use specific screening policies to respond to sexual assault allegations. However, information collected from medium-sized cities and police departments without specialized sexual assault units or specific screening policies is lacking. As the majority of police departments in the United States serve small to medium-sized populations and most do not have specialized sexual assault units, the current study advances what is known about sexual assault case processing within jurisdictions with similar characteristics. Extant research has confirmed that a combination of legally relevant and extralegal factors influences police and prosecutorial decisions in sexual assault cases. However, research on the convergence of police and prosecutors’ decisions in sexual assault cases is extremely limited (Alderden & Ullman, 2012; Holleran et al., 2010; Pattavina et al., 2016; Spohn & Tellis, 2018), especially with regard to studies that have incorporated the focal concerns perspective (Holleran et al., 2010; Spohn & Tellis, 2018). The authors of previous studies argue that it is imperative to further examine which factors overlap between police and prosecutors as this information yields a clearer understanding of sexual assault case attrition and processing.
The current study contributes to the existing body of research by assessing the congruence between the charging decisions of the police and prosecutors from a focal concerns perspective. Furthermore, the current research departs from prior studies by focusing on a medium-sized city without the presence of a specialized sexual assault unit in its police department. Eleven years of adult sexual assault cases reported to the police in a Midwestern city were analyzed to determine which factors predicted congruent charging decisions between the police and prosecutors in referred cases, regardless of whether a suspect was arrested prior to prosecutorial review. In addition, this study also analyzed the factors that were predictive of arrest. Specifically, the current study addresses the following research questions:
Research Design and Method
The research site for this study was a Midwestern city in the United States. At the time of data collection, the city had a population of nearly 110,000 people (U.S. Census Bureau, 2010). According to 2010 Census figures, the racial composition of the city was 92.4% White, 2.5% African American, 2.1% Hispanic or Latino, and 3% other races (including American Indian, Asian, and two or more races). The city has a mixed economy (including retail and manufacturing jobs) and can be described as a university town as it has several universities and community colleges in the surrounding area. The police department from which the data were obtained employed 145 sworn officers at the time of data collection. Any sworn police officer at any rank in this department was considered qualified to respond to sexual assault incidents as this department did not have a specialized sexual assault task force or unit in place. Most often, the officer who initially responded to a sexual assault claim also became the investigating officer for that case.
Data for the Study
Data for this study include police reports and court records from all adult sexual assault incidents that were reported to the police between January 1, 2000, and December 31, 2010. These reports contained information about the victim, suspect, and incident, primarily through initial and follow-up event summaries written by investigating officers. Corresponding court documents were also included for cases that were referred for prosecution. A total of 570 sexual assault incident reports were collected, encompassing 11 years of adult sexual assaults that were reported to the police department.
As this study investigated the decisions made by both the police and the prosecutors, the analyses focused exclusively on adult sexual assault cases that were forwarded to prosecutors for consideration after the police had identified a charging classification, regardless of whether a suspect was arrested prior to referral. Due to the small number of cases containing multiple offenders, female suspects, and male victims, data for this study were limited to single-offender cases where the suspect was an identified male and the victim was a female. Cases that were dismissed or considered inactive by the police (n = 307), including cases where the victim withdrew participation from the investigation (n = 178), were excluded and reserved for future analysis as the current study sought to assess decisions made by both the police and prosecutors in sexual assaults involving victims who were willing to proceed with criminal justice action. Of the 263 cases that fit these criteria, 32 cases were still pending or were missing information about the prosecutors’ charging decisions and were excluded from the data. The final sample used for this study included 231 cases. Figure 1 displays the case dispositions for the full sample. Based on their relevance in previous research, the current study examined legal and extralegal variables to determine which predict the agreement between the police classification and prosecutors’ charging decisions in sexual assault cases. The same variables were then analyzed within the context of police arrest decisions to determine which factors are influential when deciding to arrest identified suspects.
Disposition of adult sexual assault cases, 2000–2010.
Dependent and Independent Variables
The current study included two dependent variables to examine sexual assault case-processing decisions made by the police and prosecutors. The first dependent variable examined the agreement between the way the police classify or charge a felonious sexual assault on the incident report and the charging decisions of prosecutors. Examples of police classifications/charges and prosecutor’s charges include gross sexual imposition (i.e., the unwanted sexual act that was carried out), sexual assault of an adult involving force or threat of force, gross sexual imposition involving sodomy, gross sexual imposition with force, gross sexual imposition/rape with injury, and so on. The dependent variable, congruent charges, was dichotomous and coded as 1 = the police and prosecutors agree on the charge, and 0 = the police and prosecutors do not agree on the charge. Cases that were dropped by the prosecutor and cases in which the prosecutor filed a lesser charge (i.e., reduced to misdemeanor charges, such as simple assault or disorderly conduct) were included in the “do not agree” category as the charges were divergent from the police classification. The second dependent variable represented whether the police arrested an identified suspect. In this study, 90 arrests were made by the police and nearly all arrests (83/90) occurred before the cases were referred to prosecutors. Arrest was dichotomous and was coded so that 0 = no arrest and 1 = arrest.
Based on prior research, and drawing from the focal concerns perspective and Estrich’s (1987) concepts of differential case processing for “real rapes” and “genuine victims,” the independent variables in the current study are organized within four groups: victim and suspect demographics, evidentiary factors, seriousness of the crime and dangerousness of the offender, and blame and believability of the victim. Because the focal concerns theory posits that a perceptual shorthand includes demographic characteristics, such as age, race, and gender, victim and suspect demographics were incorporated and included the age of the victim and suspect and the race of the suspect. Victim and suspect age were continuous variables, with a victim age range of 18–80 (M = 28.05, Mdn = 24) and a suspect age range of 18–77 (M = 35.47, Mdn = 32). The race/ethnicity of victims was identified in only 45 of the cases in this study, and only six victims were described as non-White, therefore, victim race could not be assessed in the current study. Suspect race was dichotomized as 0 = White and 1 = non-White.
Conceptually similar to Spohn and Tellis (2018), the evidentiary factors in the current study served as controls based on their importance in case-processing decisions. In this study, evidentiary factors included three legally relevant variables: the collection of physical or forensic evidence, the presence of witnesses, and whether the victim completed a rape kit (i.e., sexual assault forensic exam). Whether physical or forensic evidence was obtained from the scene was dichotomized as 0 = no and 1 = yes. Evidence included forensic evidence, such as bedding, clothing, hair, or other objects believed to contain DNA that were obtained by investigators at the scene or were presented to the police by the victim or a third party. Evidence also included physical evidence, such as video tapes with recorded footage of the incident, empty condom wrappers, and other evidence collected by or provided to investigators. The presence of witnesses who could attest to the occurrence of the incident was also dichotomized, so that 0 = no witnesses and 1 = at least one witness. The final evidentiary variable included whether the victim had completed a rape kit (0 = no, 1 = yes).
The focal concerns perspective contends that case outcomes are influenced by the seriousness of the crime and the perceived dangerousness of the offender. The current study included three predictors to assess crime seriousness and offender dangerousness: victim injury, physical resistance, and weapon use by the suspect. Sexual assaults are perceived to be more serious by the police and prosecutors when there is a physical struggle between the victim and assailant, and when the victim sustains physical injuries (Estrich, 1987). In this study, whether the victim sustained physical injuries such as bruises, cuts, or abrasions and whether the victim reported that she physically resisted the assault were each coded 0 = no and 1 = yes. Weapon use by the suspect included all weapon types (e.g., firearms, knives, and blunt instruments) and was dichotomized to reflect 0 = no weapon and 1 = weapon.
Prior research has shown that the credibility of sexual assault victims is often questioned by the police and prosecutors and victims are assigned a certain amount of blame for their assault, especially in cases lacking evidence. In the current study, blame and believability is conceptualized similar to Holleran et al. (2010) and included three measures that called into question the character and credibility of the victim. These measures were the timeliness of the report, risk-taking behavior of the victim, and the relationship between the victim and suspect. The time lapse between the incident and when it is reported is important to include as research has found police and prosecutors to be more apt to pursue sexual assault cases in which the victim promptly reported the incident. In this study, timely report was measured by assessing whether the victim reported the assault to the police within 24 hr of when it occurred (0 = more than 24 hr and 1 = within 24 hr). Risky behavior was coded 1 if the victim was engaged in any of the following during the time of incident: went to a bar, lounge, or club prior to the incident; was hitchhiking; or was using drugs or alcohol. The relationship between the victim and suspect was coded to reflect whether the suspect was a stranger to the victim, an acquaintance, or an intimate partner and, similar to Holleran et al. (2010) and Spohn and Tellis (2018), was coded so that stranger was the reference category.
Findings
The data from the police reports and court documents were used to examine three research questions: (a) To what extent are the crime classifications listed by the police on the incident report congruent with the prosecutors’ charging decisions in adult sexual assault cases? (b) Which factors predict the agreement between the police classification and prosecutors’ charging decisions in sexual assault cases? and (c) Are the same factors that influence congruent charging decisions also predictive of police decisions to arrest?
To address the first research question, descriptive statistics explored the dispositions of the 231 adult sexual assault cases in which charging decisions were made by both the police and prosecutors. Frequency distributions illustrated that the police classification of the crime and prosecutor’s charging decisions were in agreement in approximately one third of the sexual assault cases (79/231 or 34%). Figure 1 illustrates that the remaining two thirds were removed from the criminal justice funnel through dismissals (134/231 or 58%) or charge reductions (18/231 or 8%). In addition to providing the frequency and percentage for each categorical independent variable among all 231 cases, Table 1 illustrates the breakdown of the characteristics of sexual assault cases with congruent and incongruent charging decisions between the police and prosecutors and characteristics of cases in which the police made an arrest. A few notable findings were initially observed from this analysis. First, charging decisions were congruent in a higher percentage of cases in which victims incurred physical injuries. Charging decisions were congruent in 31.1% of the cases in which the victim sustained physical injuries, whereas 18.8% of incongruent cases involved physical injuries to the victim. In addition, a victim physically resisted the attack in 30.3% of congruent cases, whereas 14.8% of incongruent cases involved physical resistance. The majority of the sexual assault cases (74.5%) were reported to the police within 24 hr. Further observation revealed that timely reporting occurred in a slightly higher percentage of congruent cases than incongruent cases. Similarly, 87.8% of cases where the police arrested an identified suspect were reported within 24 hr. Finally, a weapon was used in 28.6% of cases that resulted in congruent charges and in 42.4% of incongruent cases.
Descriptive Statistics and Case Outcomes for All Categorical Variables.
Note. *For all cases—Median victim age = 24 (range 18–80); median suspect age = 32 (range 18–77).
To answer the second and third research questions, two models were constructed (charging agreement and arrest) and evaluated using multivariate analyses. Because the dependent variables were dichotomous, binary logistic regression models were used to determine the factors that predicted congruent police and prosecutor charging decisions and police arrest decisions. Prior to the analysis, model diagnostics were conducted to screen for multicollinearity and to ensure that the models were a good fit for the data (Midi et al., 2010). No multicollinearity issues were detected as all tolerance scores were over 0.70 and all variance inflation factor (VIF) scores were below 2. The Hosmer and Lemeshow goodness-of-fit test for each model yielded a statistic that was not significant, indicating both models were appropriate for the data.
The results of the multivariate analysis on charging agreement are presented in Table 2. The analysis revealed four variables to be significant predictors of charging agreement between the prosecutors and police. None of the victim or suspect demographic variables were significantly related to charging agreement; rather, significant predictors were found within each of the other three categories of variables (i.e., evidentiary factors, seriousness of the crime, and blame and believability). The most robust predictor of charging agreement was the presence of physical or forensic evidence. Cases containing physical or forensic evidence were 7.481 times more likely to result in congruent charges than cases without evidence. In addition, cases in which the victim had completed a rape kit were 3.175 times more likely to result in congruent charges than cases in which a rape kit was not collected. Physical resistance by the victim was also associated with congruent charges. Conceptualized in this study as an indicator of case seriousness, sexual assaults in which the victim physically resisted the suspect were 2.298 times more likely to result in congruent charges than cases in which the victim did not physically resist.
Logistic Regression Results for Charging and Arrest Decisions.
Note. *p < .05. **p < .01. ***p < .001.
One blame and believability factor was also found to be predictive of charging agreement. Timely reporting of the incident was associated with congruent charging decisions. Charging decisions in sexual assault incidents that were reported to the police within 24 hr were 3.172 times more likely to be congruent than incidents that were reported more than 24 hr after they occurred. Charging outcomes, however, were not affected by the relationship between the victim and suspect, risk-taking behaviors of the victim, weapon use by the suspect, victim injuries, or the presence of witnesses.
Turning to police arrest decisions, Table 2 illustrates that six predictors emerged as significant in the model. Similar to the charging agreement model, the arrest of an identified suspect was influenced by a combination of legal and extralegal factors. Predictors of arrest were significant in all four categories of variables. Whereas none of the victim or suspect demographics were influential in charging decisions at the prosecutorial stage, the age of the suspect was positively related to the decision to arrest. The odds of arrest increased by a factor or 1.033 for each year of suspect age.
Similar to charging agreement, the presence of physical or forensic evidence was the most salient predictor of arrest. Cases with physical or forensic evidence were 13.745 times more likely to result in the arrest of an identified suspect. Also similar to its impact in charging decisions, the completion of a rape kit increased the odds of an arrest by a factor of 2.555. In contrast to prosecutorial charging decisions, physical resistance by the victim was not associated with arrest outcomes. However, the other two indicators of case seriousness were significant predictors of arrest, neither of which influenced charging decisions at the prosecutorial level. Specifically, cases in which the victim suffered physical injuries were 2.959 times more likely to result in arrest and suspects who used a weapon were 1.832 times more likely to be arrested. The blame and believability variable that significantly predicted charging agreement was also significantly predictive of arrest. Sexual assault cases that were reported within 24 hr were 2.876 times more likely to result in an arrest. Consistent with the charging agreement results, risk-taking behavior by the victim, the relationship between the victim and suspect, and the presence of witnesses were not significant predictors of arrest.
In sum, three of the four variables that were significant predictors of charging agreement were also predictive of arrest decisions. Physical resistance by the victim was significant at the prosecutorial stage but did not affect arrest decisions. Conversely, suspect age and two indicators of crime seriousness and offender dangerousness were associated with the likelihood of arrest (weapon use by the suspect and physical injuries to the victim) but were not significantly related to charging agreement. Moreover, both stages of case processing were influenced by the strength of evidence, the seriousness of the case, and victim credibility factors.
Discussion and Conclusion
The purpose of this study was to examine the factors that contributed to congruent charges between police and prosecutors in sexual assault cases involving victims who were willing to proceed with criminal justice action and to determine which factors were predictive of arrest. Of particular interest was the degree to which the focal concerns perspective could be applied to charging decisions made by the police and prosecutors and arrest decisions of the police. The focal concerns perspective (as it relates to police and prosecutors) posits that the seriousness of the offense, the blameworthiness of the offender, and the likelihood that the case will result in a conviction underlie the decisions made by the police and prosecutors in sexual assault cases. According to this perspective, police and prosecutors incorporate both legal and extralegal factors into their decision-making processes. Consistent with the focal concerns perspective and previous studies, this study found that the decisions made by the police and prosecutors were influenced by a combination of legal and extralegal factors.
The most robust predictor of both arrest and charging agreement was the collection of physical or forensic evidence from the scene of the offense. This finding was consistent with prior research and the focal concerns perspective as the presence of evidence significantly increases the certainty that the crime occurred (Holleran et al., 2010). Unique from Holleran et al. (2010) and Spohn and Tellis (2018), the current study assessed the impact of the completion of a rape kit. Operationalized to represent an evidentiary source, this factor was a significant predictor of both arrest and charging congruency. As the forensic results were unknown in most cases in which a rape kit was completed, this finding could also reflect victim credibility. The significance of a rape kit implies that the police and prosecutors might have anticipated that the results from the rape kit would corroborate the victims’ accounts and would be used as evidence in court. However, it is also possible that the police and prosecutors were less likely to doubt victims’ claims when victims were willing to complete a rape kit.
Indicators of crime seriousness and offender dangerousness were also significantly related to the decision to arrest and congruent charges; however, the specific indicators that were significant were different at each stage. Arrests were more likely in sexual assaults in which the victim sustained physical injuries and in sexual assaults that involved a weapon; whereas, charges were more likely to be congruent in cases that involved physical resistance from the victim. From the focal concerns perspective, arresting a dangerous offender is necessary to protect the victim and others, and police in this study may have used arrest to remove an immediate threat from society. In addition, weapon involvement and victim injury are, in general, objective factors that offer police probable cause to make an arrest. That these two variables failed to reach significance in charging agreement at the prosecutorial stage was unexpected and oppose the results from prior research that has found victim resistance influential to arrest and victim injury predictive of felony charging decisions (Alderden & Ullman, 2012). Instead, the current study found charging agreement was more likely in sexual assault cases where the victim physically fought the suspect. The significance of physical resistance by the victim is unique from previous research that has examined sexual assault cases from a focal concerns perspective (Holleran et al., 2010; Spohn & Tellis, 2018). It could be that the prosecutors in this study predicted a higher likelihood of convictions when sexual assault cases involved physical resistance from the victim as this variable focuses on the actions of both the suspect and victim. In other words, suspects in sexual assaults are perceived as being more dangerous when there is a physical confrontation and, by struggling with the suspect, the victim has clearly shown they do not consent to the suspect’s actions. It is also important to note that the data in this study were reflective of the pre-2013 FBI (Federal Bureau of Investigation) definition of rape that strongly emphasized the use of force during the commission of the crime. It is possible that, in the current study, incidents involving victims who physically resisted fell in line with the federal definition more clearly, making convictions more likely in the eyes of prosecutors.
One indicator of victim blame and believability was significantly related to arrest and congruent charging decisions. Similar to previous research, sexual assaults that were reported promptly (i.e., within 24 hr) were more likely to lead to an arrest and to congruent charges (Spohn & Tellis, 2018; Tasca et al., 2013). This finding contrasts with Holleran et al.’s (2010) findings as they found that timely reporting was not a significant predictor of charging agreement in sexual assaults where an arrest had been made prior to the case referral to prosecution. However, Spohn and Tellis (2018) found timely reporting to be a significant predictor of arrest and case referral to prosecutors, including cases in which a suspect had not yet been arrested. In Spohn and Tellis (2018), a timely report was defined as an incident that was reported within 1 hr, versus the 24-hr threshold used in the current study. Worth noting is the importance of considering the link between prompt reporting and other case factors. In other words, as the length of time between the incident and reporting the incident to the police increases, the more difficult it may be to collect physical or forensic evidence, and the more likely that physical injuries will be less evident. Risk-taking behavior by the victim and the relationship between the victim and suspect did not influence arrest or charging congruency in the current study. These findings are distinct from Holleran et al.’s (2010) analyses as their study revealed that congruent charges were more likely to be filed in sexual assault cases that involved strangers and those in which the victim was not involved in risky behavior preceding the incident.
The results from this study showed that the variables that significantly influenced prosecutors’ charging decisions and the variables influential in arrest decisions were highly similar. These findings support the focal concerns perspective and Frohmann’s (1997) assertion of a downstream orientation, whereby prosecutors attempt to predict case outcomes and grant preference to cases perceived as leading to convictions. Just as prosecutors attempt to make predictions about how a jury will perceive a case, the police make predictions about how the case will be handled by the prosecutors. Police officers in this study were aware that sexual assault cases with overwhelming physical and forensic evidence were more readily accepted by prosecutors. Consequently, sexual assault cases that contained sufficient evidence in the eyes of the police, including legally relevant factors, such as evidence, weapon use, and victim injury, were forwarded to the prosecutors for their review. It appears that the police in this study shifted any uncertainty they had about sexual assault cases to prosecutors as only 34% (79/231) of the sexual assault cases referred to prosecutors were accepted and charged with the same felony offense listed in the police classification. To elaborate further, in this study, an arrest was made in only 90 sexual assault cases; 83 out of 90 (92%) arrests were made prior to case referral and, of those 83 cases, 72 (87%) resulted in congruent charges. This implies that there is indeed a downstream orientation and, similar to prior research, that prosecutors are likely weighing in at the prearrest stage (Spohn & Tellis, 2018). Along with the attempt to avoid uncertainty, the police in this study may have been reluctant to arrest suspects prior to prosecutorial review as they may have predicted that most cases would be rejected by prosecutors, which would ultimately affect the number of cases cleared by arrest. As very few arrests were made after cases were sent to prosecutors for review, it appears that the police in the current study arrested suspects in cases that contained profound evidence, while reserving questionable cases for prosecutors to evaluate. Spohn and Tellis (2018) noted that sexual assault cases were often sent to prosecutors for evaluation as a means of removing cases that were viewed as weak from further processing. The same may hold true in this study, given the high level of attrition at the referral stage.
As with all research, the current study has a number of limitations. First, this study was constrained by the quality of the archival records from which the data were collected. It is possible that key factors that guided the decisions made by police and prosecutors were not captured from the documents used for this study. In addition, the narratives in the police reports may not accurately reflect how the sexual assault cases were handled by the police officer or any biases held by the police and prosecutors. Second, the results from this study may not be generalizable as the police reports and court documents were gleaned from one Midwestern police agency that does not have a specialized unit for sexual assault cases. Replications in other jurisdictions with and without specialized units may or may not produce similar results. Similarly, the racial composition of the city from which these data were obtained was racially homogeneous. Differing results may be found in more diverse jurisdictions. In a similar vein, the race of the victim was missing in most cases, so this study could not examine race in relation to the mixed results of victim race and victim/suspect racial dyad that has been found in recent research (O’Neal et al., 2016, but see also Spohn & Tellis, 2018). Third, of the 570 total cases, this study focused on the 231 cases that were forwarded to prosecutors by the police and had a known prosecutorial outcome listed. As such, this study did not examine why the remaining cases were not processed further. It is possible that victims who were engaged in multiple forms of risk-taking behavior either did not report the sexual assault to the police or those cases were not presented to the prosecution. Analyses on case attrition at these stages are currently underway. Finally, the current study did not assess the potential influence of individual officer and prosecutor characteristics to determine whether these factors affected the way reports of sexual assault were handled. Although the current study sought to examine the processing of sexual assaults from a systems perspective, individual officer and prosecutor characteristics may provide valuable information about the way sexual assaults are handled. Despite these limitations, this study provided insight into the focal concerns of the police and prosecutors for adult sexual assault cases in a Midwestern city.
In conclusion, this study confirmed the unfortunate reality of high attrition in sexual assault cases. Over half of the 570 sexual assault incidents in this research were filtered out of the criminal justice system at the point of police contact. Of the remaining incidents, prosecutors agreed to felony charges in less than one third of the cases that they were presented. Finally, only 45 incidents were processed to the fullest extent and resulted in guilty as charged verdicts. Taken together, this means that only 8% of the adult sexual assault cases that were reported to the police in this study resulted in a felony conviction. This is troubling, as the victims in the current study were willing to cooperate with the police and prosecutors, yet most were met with resistance at both points of contact. In addition, consistent with extant literature, most of the sexual assaults in this study did not conform to the stereotype of “real rapes.” Table 1 highlighted that 83.8% of the cases did not contain evidence, 71% did not have witnesses, 77.1% of victims did not sustain physical injuries, 79.6% of victims did not physically resist, 62.5% had no weapon involvement, and 70.1% of victims were engaged in risk-taking behavior.
Furthermore, this study emphasized that both legal and extralegal factors indeed influence sexual assault case processing. Aligned with the focal concerns perspective, evidentiary sources provided the primary motivation for case processing, followed by extralegal factors that center on the credibility of the victim. Sexual assault cases that were predicted to lead to convictions were pursued further by police and prosecutors, whereas cases lacking these features were funneled out by way of dismissals or reduced charges. In seeking winnable cases, the police and prosecutors in this study focused heavily on evidentiary sources to avoid any uncertainty about the case. Although risk-taking behavior of the victim was not itself found to be predictive of arrest or congruent charges, an indicator of blame and believability (i.e., timely reporting) was influential, and slightly more so during the prosecutorial stage. These findings support other research that has shown that prosecutorial concerns about whether a case would lead to a conviction might result in prosecutors incorporating the credibility of the victim into their decision-making (Alderden & Ullman, 2012; Frohmann, 1997).
The results of this study also highlight the interactive relationship between the police and prosecutors in sexual assault cases. The police are commonly considered the gatekeepers of the criminal justice system as their decisions affect the decisions at each subsequent stage. However, with regard to sexual assault, it appears that prosecutors stand alongside the police at the gate and work in tandem with the police to determine how to process sexual assault incidents. Arrests in this study were made prereferral in cases with profound evidence, whereas cases that were less than certain were referred for consideration before the police devoted additional investigative resources to them. Future research should continue to examine the police–prosecutorial junction in sexual assault cases to determine whether the results from this study can be applied in other areas.
A few implications arise from the results of this study. First, a major focus of prosecutors and ultimately the police in sexual assault cases is to pursue cases that are perceived to be “slam-dunk” cases in the courtroom. As such, one implication is that police agencies include specialized sexual assault units with officers who have been extensively trained in how to properly interview traumatized victims, interrogate suspects, and collect quality evidence. In addition, victim advocates should be included to provide support and information to victims and to assist the police in gathering information (e.g., by explaining to the victim matters such as why the police are asking sensitive questions). Using a victim-centered approach where victims are not met with skepticism may encourage victims to report incidents of sexual assault and to continue with case processing.
Along with police training and the inclusion of specialized sexual assault units, prosecutors should accept cases that meet the legal standard of the crime, rather than being guided by trial sufficiency standards, and prosecutors and judges should convey to jurors the legal standards from which to make their determinations. In general, the police and prosecutors are well-aware of the most common characteristics in sexual assault cases, and that those characteristics are not aligned with the original concept of a “real rape.” Therefore, a reconceptualization of what a “real rape” entails is necessary and should be reinforced among the police and prosecutors. In addition, informing jurors and the general public about what a typical “real rape” looks like may help to reduce the number of rape myths that are widely adopted by society. Recent media attention that has been drawn to sexual assault cases involving prominent suspects may open the door for this discussion in broader society. Further examination and replication is needed to assess whether the findings of this study are unique to the jurisdiction used.
Footnotes
Declaration of Conflicting Interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The author(s) received no financial support for the research, authorship, and/or publication of this article.
