The Sexual Stratification Hypothesis: Is the Decision to Arrest Influenced by the Victim/Suspect Racial/Ethnic Dyad?

Abstract

The sexual stratification hypothesis suggests that criminal justice responses to sexual victimization will differ depending on the victim/suspect racial/ethnic dyad. Previous research examining the sexual stratification hypothesis has primarily focused on court processes, and the small body of literature examining arrest decisions is dated. There remains substantial opportunity for testing the sexual stratification hypothesis at response stages apart from the court level (i.e., arrest). Using quantitative data on 655 sexual assault complaints that were reported to the Los Angeles County Sherriff’s Department (LASD) and the Los Angeles Police Department (LAPD) in 2008, this study examines the effect of the victim/suspect racial/ethnic dyad on the decision to arrest. Findings suggest that police consider the victim/suspect racial/ethnic dyad when making arrest decisions. In addition, victim characteristics, strength of evidence indicators, and measures of case factors predict the police decision to make an arrest.

Keywords

sexual stratification hypothesis sexual assault case processing police arrest decisions race ethnicity

Criminal justice decision making and official responses to sexual assault have been shaped by the justice system’s complicated history regarding complex racial politics (Belknap, 2007; Kennedy, 1997; LaFree, 1980; Miller, 2008; Wolfgang & Reidel, 1973). In fact, Brownmiller (1975) notes, “No single event ticks off America’s political schizophrenia with greater certainty than the case of a black man accused of raping a white woman” (p. 230). The salient role of the victim/suspect racial/ethnic dyad in sexual assault case processing originated from the use of Black-on-White rape accusations to justify social control of Black males and to ensure White male dominance in the wake of emancipation (Kennedy, 1997; Patton & Snyder-Yuly, 2007).

Although research has established that the victim/suspect racial/ethnic dyad influences sexual assault case outcomes (see Spohn & Spears, 1996; Walsh, 1987), scholarly work has primarily focused on examining prosecutorial and judicial decision making (Beichner & Spohn, 2005; Kingsnorth, Lopez, Wentworth, & Cummings, 1998; Kingsnorth, MacIntosh, Berdahl, Blades, & Rossi, 2001; Spohn & Holleran, 2001; Spohn & Spears, 1996; Tellis & Spohn, 2008¹; Walsh, 1987). Only a handful of studies on sexual assault case processing have examined earlier decision points such as the decision to arrest (see Alderden & Ullman, 2012a, 2012b; Bouffard, 2000; Tasca, Rodriguez, Spohn, & Koss, 2012). Even fewer studies examining sexual assault case processing have specifically investigated how the racial and ethnic composition of the victim/suspect dyad influences arrest decisions, and the studies that have are largely outdated (e.g., LaFree, 1980, 1981).

This is a notable oversight considering police officer decision making plays a crucial role in determining sexual assault case outcomes. After all, the police have been called “gatekeepers” of the criminal justice system because they determine whether persons who are accused of sexual assault will continue to later stages of the criminal justice process (Kerstetter, 1990). Law-enforcement officers have discretionary powers that include deciding whether to record an incident report, how a call for service will be handled, how much time will be devoted to investigating a case, the incident type to formally record, and whether the suspect will be arrested. Because of these discretionary powers, some scholars argue that examining arrest decisions—as opposed to later stages of the criminal justice process—is most important because doing so helps fully understand the factors that affect sexual assault case processing (Alderden & Ullman, 2012a).

Accordingly, we draw on the sexual stratification hypothesis to examine the effect of the victim/suspect racial/ethnic dyad on the decision to arrest. By doing so, we can assess whether the considerable amount of discretion bestowed on police officers facilitates their reliance on extra-legal factors (including race and ethnicity) when deciding whether or not to make a sexual assault arrest. To address this concern, we use quantitative data on 655 sexual assault complaints that were reported to the Los Angeles County Sherriff’s Department (LASD) and the Los Angeles Police Department (LAPD) in 2008.

Literature Review

The Sexual Stratification Hypothesis

The sexual stratification hypothesis is a variation of conflict theory that was originally developed to explain racial disparities in both criminal justice decision making and responses to victimization and was later extended to include ethnic disparities (see Tellis & Spohn, 2008). Conflict theory views social control as a means to advance and protect authority power and interest, rather than as a response to practical concerns such as an increase in crime rates (Liska, 1992). The sexual stratification hypothesis specifically posits that persons of influence control sexual access according to race and ethnicity (Tellis & Spohn, 2008). Regarding sexual assault, Black and Hispanic men who assault White women can be viewed as disrupting the power hierarchy of the dominant group (i.e., White men). Therefore, these types of assaults will be punished more harshly than sexual assaults involving other victim/suspect racial/ethnic dyads (Tellis & Spohn, 2008). In addition, Black and Hispanic men who assault Black and Hispanic women will be treated more leniently by the justice system because the victimization of these women is viewed as less serious and less important (Tellis & Spohn, 2008).

The American legal system’s failure to extend equal protection to women of color who are sexually assaulted can be linked back to the historical belief that the obedience of Blacks can only be obtained and maintained through “uncontrolled authority over the body” (see Kennedy, 1997, p. 33). As a result, Black-on-White sexual assault is believed to be more serious than White-on-White sexual assault and White-on-White sexual assault is seen as more serious compared with Black-on-Black sexual assault (LaFree, 1980; Wolfgang & Reidel, 1973). These beliefs are often held by those who benefit from race/power differentials (Walsh, 1987) and wish to maintain societal power imbalances (i.e., Whites). LaFree (1980) states, “American society is characterized by a sexual stratification system which imposes more serious sanctions on men from less powerful groups who are accused of assaulting women from more powerful social groups” (p. 852). Consistent with this reasoning, the sexual stratification hypothesis argues that Whites who assault Black or Hispanic individuals will receive the most lenient punishments. In addition, Blacks or Hispanics who assault Black or Hispanic individuals will receive lenient punishments. Conversely, Blacks or Hispanics who assault Whites will receive the harshest sanctions (Tellis & Spohn, 2008; Walsh, 1987). As a recent example, Sommers, Goldstein, and Baskin (2014), in their study of prosecutorial decision making and violent crime, found that prosecutors were less likely to make plea offers in cases involving suspects of color (i.e., Blacks and Hispanics) and White victims.

Sexual assault is a relational phenomenon that happens in the context of a relationship between two people. This makes the sexual stratification hypothesis an appropriate framework to examine racial and ethnic disparities in criminal justice decision making and responses to victimization.² This framework is appropriate because it examines the relationship between the victim and suspect by examining racial and ethnic dyads as a single variable as opposed to looking at victim race/ethnicity and suspect race/ethnicity as two separate variables. In the following section, we turn to prior research on racial and ethnic disparities in police decision making for sexual assault cases. Because empirical studies examining the victim/suspect racial/ethnic dyad are scarce for arrest outcomes in sexual assault cases, the following section also includes prior research on race and ethnicity generally for the overall policing stage.

Race, Ethnicity, and Police Decision Making in Sexual Assault Cases

There are multiple decision points in sexual assault investigations where race, ethnicity, and a number of other extra-legal characteristics can influence case attrition (Tasca et al., 2012). The initial step required to prompt a police investigation is the victim report. The police can influence whether the victim files a report after initial contact—either through persuasion or a refusal to take a report (Alderden & Ullman, 2012a; Campbell, 2005, 2006; Maier, 2008). Factors that predict victim reporting include the race and ethnicity of the victim and suspect, victim injury, victim resistance, presence of a weapon, and the victim/suspect relationship (Bachman, 1998; Du Mont, Miller, & Myhr, 2003; Felson & Pare, 2005; B. S. Fisher, Daigle, Cullen, & Turner, 2003; Rennison, 2002). Findings have been mixed regarding race, ethnicity, and likelihood of reporting. Some studies have found that White victims are more likely to report a sexual assault compared with victims of color (Tjaden & Thoennes, 1998). Other studies have found the opposite (Bachman, 1998; Felson & Pare, 2005; B. S. Fisher et al., 2003). Few studies on victim reporting have examined the influence of the victim/suspect racial/ethnic dyad, yet B. S. Fisher and colleagues (2003) did find that victims were more likely to report a sexual assault when the victim and suspect race differed.

Once an incident has been reported, police officers determine whether the case meets the criteria for sexual assault and decide whether to found or unfound (dismiss) a case (Alderden & Ullman, 2012b; Kerstetter, 1990). If a case is founded, it remains open for investigation and police proceed to questioning witnesses, identifying the suspect, and collecting evidence (Alderden & Ullman, 2012b; Frazier & Haney, 1996; Kerstetter, 1990; Shaw & Campbell, 2013; Tasca et al., 2012). Similar to victim reporting, findings regarding the relationship between race/ethnicity and sexual assault investigations have been inconsistent. Bryden and Lengnick’s (1997) review concluded that Black victims were less likely to have their cases founded, while Alderden and Ullman (2012a) reported no such relationship. Mixed results on this decision stage have also been documented when taking the victim/suspect racial/ethnic dyad into account (Bouffard, 2000; Tellis & Spohn, 2008). Regarding investigation practices, Tasca and colleagues (2012) did not find any differences for suspect identification according to the victim/suspect racial/ethnic dyad, but Shaw and Campbell (2013) found that police officers were actually more likely to submit sexual assault kits for non-White victims.

Overall, research on the earliest policing decision points of sexual assault case processing reveals varied support for the influence of race and ethnicity on police decisions. Instead, the main predictors of these early investigative decisions appear to revolve around matters of evidence, a prior victim/suspect relationship (increases the likelihood of founding a case and identifying a suspect),³ and questions regarding the victim’s moral character and behavior at the time of the incident (Alderden & Ullman, 2012a; Bouffard, 2000; Frazier & Haney, 1996; Kerstetter, 1990, Tellis & Spohn, 2008). Nevertheless, these findings do point to the fact that police officers rely on race and ethnicity to make decisions—at least in some instances, which indicates the possibility that race and ethnicity may factor into the police decision to arrest.

The later stages in the law enforcement component of sexual assault case processing include the arrest of the suspect and the presentation of the case to the prosecutor. In these final law enforcement stages, police and prosecutorial decisions are frequently interrelated and dependent on one another. Although these events are often linked to one another (e.g., the prosecutor issues charges, warranting the arrest of the suspect), they do not always happen together. For example, the police may refer a case to the prosecutor in efforts of obtaining a warrant, but the prosecutor may deny an arrest warrant. Then again, it is also possible for the police to make an arrest prior to referring the case to the prosecutor and subsequently release the suspect after making the decision to not refer the case to the prosecutor. Overall, there are numerous possible case outcomes including no arrest and no referral to the prosecutor, only an arrest, only a referral to the prosecutor, or both an arrest and a referral to the prosecutor. It should be noted that for a case to move forward in the criminal justice system from the investigation stage to the prosecution stage, there must be both an arrest and a referral. In our “Method” section, we provide detailed information on how we handle and code the arrest decision.

The present study focuses on the decision to arrest for two main reasons. First, significant attrition occurs at the arrest stage. According to the National Crime Victimization Survey, only 31% of reported sexual assaults from 2005 to 2010 resulted in arrest (Planty, Langton, Krebs, Berzofsky, & Smiley-McDonald, 2013). Therefore, the arrest decision plays an important role in shaping whether reported sexual assaults move forward in the criminal justice system. Second, statistics reveal that suspects of color—and Black suspects in particular—are overrepresented in arrests for sexual assault (Greenfield, 1997). Specifically, Black suspects comprise 42% of all arrestees for sexual assault, which is nearly 30% more than the percentage of Blacks represented in the general population (Greenfield, 1997; U.S. Census Bureau, 2010). Despite the obvious overrepresentation of suspects of color in arrests for sexual assault, only a handful of studies have investigated whether the victim/suspect racial/ethnic dyad specifically influences the likelihood of arrest when controlling for other legal and extra-legal variables (but see Bouffard, 2000; LaFree, 1980, 1981; Tasca et al., 2012).

Recent research on sexual assault case processing, however, has examined the effect of other extra-legal characteristics on arrest decisions, including detective gender, questions regarding the victim’s character and reputation, alcohol use by the victim and offender, and the victim/suspect relationship. Notably, extra-legal factors have been found to heavily influence the decision to arrest (see Alderden & Ullman, 2012a, 2012b; Bouffard, 2000; Felson & Pare, 2007; Spohn & Tellis, 2014; Tasca et al., 2012). Findings from these studies demonstrate the immense discretion that characterizes arrest for sexual assault and point to the possibility that the victim/suspect racial/ethnic dyad may also influence the decision to arrest. Indeed, the amount of crime makes it impossible for the police to enforce all law violations, allowing room for reliance on stereotypes of sexual assault and extra-legal factors in decision making (Alderden & Ullman, 2012a; S. Z. Fisher, 1993; Goldstein, 1960; Mastrofski, 2004; Reiner, 2010; Rowe, 2007).

Current Focus

The sexual stratification hypothesis suggests that criminal justice responses to victimization will differ depending on the victim/suspect racial/ethnic dyad (LaFree, 1989). Previous research examining the sexual stratification hypothesis, however, has resulted in mixed findings (see Spohn & Tellis, 2012). As mentioned above, there remains substantial opportunity for testing the sexual stratification hypothesis at response stages apart from the court level (i.e., at arrest). The present study addresses issues related to this goal and contributes to the current body of literature in three important ways. First, this study tests the sexual stratification hypothesis by examining arrest decisions. Testing the hypothesis in a context other than court processes contributes to extant research and extends the hypothesis’ application. Second, this research examines theoretically relevant variables regarding factors that influence sexual assault case outcomes. Third, we expand the early framework that primarily focused on Black and White victims and suspects to include other individuals of color. We move from a focus on just Black and White individuals to Black, White, and Hispanic suspects and victims. Previous research on racial disparities in criminal justice decision making, the criminal justice system’s unequal enforcement of the law, the American legal system’s failure to protect Blacks from criminality, sexual assault case outcomes, and research testing the sexual stratification hypothesis inform the analytic strategy below.

Method

Data

We analyze quantitative data on sexual assault complaints that were reported to the LASD and the LAPD in 2008, which were collected for a study on the policing and prosecuting of sexual assault (Spohn & Tellis, 2012). For the original study, 944 case files were obtained for sexual assaults with female victims above the age of 12 that were reported in 2008.⁴ We obtained data on the following sex crimes: rape, attempted rape, sexual penetration with a foreign object, oral copulation, sodomy, unlawful sex, and sexual battery. From the LASD, case files were obtained for all reports that met the selection criteria above (N = 543). Due to the numerous cases reported to the LAPD in 2008, a stratified random sample of cases was selected (N = 401). Because we wanted to ensure an adequate number of cases from each of the LAPD’s 19 divisions, as well as an adequate number of cases from each case clearance category (cleared by arrest, cleared by exceptional means, investigation continuing, and unfounded), the sample was stratified by LAPD division and, within each division, by the type of case clearance (for more information on the data and collection, see Spohn & Tellis, 2014). For the purpose of the present study, we eliminated all cases in which there was an unidentified suspect; the police obviously cannot make an arrest in cases with unidentified suspects. We also eliminate all unfounded cases. If the investigating officer believes the victim’s account and determines that the reported incident constitutes a crime, the case becomes one of the “crimes known to the police” that will be included in the jurisdiction’s crime statistics. If, however, the officer does not believe the victim’s story and therefore concludes that a crime did not occur, the case is unfounded. After imposing these restrictions on the original sample, 655 cases met the selection criteria and were included in the current study.

Dependent Variable

The present study focuses on the police decision to arrest (1 = yes, 0 = no). Specific to our study, it is necessary to point out that our determination of whether the suspect was arrested (or not) is not based on whether the case was cleared by arrest for Uniform Crime Reporting (UCR) purposes.⁵ This reflects the fact that Spohn and Tellis (2014) found that official data on cases that were “cleared by arrest” were misleading due to the fact that both the LASD and (especially) the LAPD changed the case clearance from cleared by arrest to cleared by exceptional means when the district attorney refused to file charges. Our measure of arrest thus differentiates between cases in which the police made an arrest (regardless of how the case was cleared) and cases in which a suspect was not arrested. As shown in Table 1, approximately 53.6% of the suspects were arrested (n = 351). It should be noted that eliminating the unfounded cases and cases without an identified suspect inflated the percentage of suspects arrested. In the original sample of 944 cases, 39.6% of the reports resulted in the arrest of the suspect.

Table 1.

Policing Sexual Assault, LAPD and LASD, 2008 (N = 655).

Variables	M or %
Dependent variable
Police arrested the suspect	53.6%
Independent variable
Suspect Black/victim Black	16.3%
Suspect Black/victim Hispanic	4.3%
Suspect Black/victim White	4.3%
Suspect Hispanic/victim Black	2.0%
Suspect Hispanic/victim Hispanic	42.0%
Suspect Hispanic/victim White	9.8%
Suspect White/victim Black	0.9%
Suspect White/victim Hispanic	4.6%
Suspect White/victim White (reference)	15.9%
Control variables
Los Angeles Police Department	35.4%
Suspect/victim characteristics
Intimate	27.1%
Nonstranger	53.8%
Stranger (reference)	19.1%
Victim age	25.76 (SD = 12.26)
Suspect age	32.24 (SD = 11.84)
Victim consumed alcohol	23.3%
Victim’s credibility questioned	6.0%
Strength of evidence
Physical evidence collected	41.2%
Suspect interviewed	59.2%
Victim cooperated	74.1%
At least one witness	46.0%
Prompt reporting	24.7%
Case factors
Aggravated sexual assault	57.6%
Suspect physically assaulted victim	45.2%
Victim resisted	75.7%

Note. LAPD = Los Angeles Police Department; LASD = Los Angeles County Sherriff’s Department.

Independent Variable of Interest

Table 1 also presents summary statistics on the independent variable and control variables. The primary independent variable of interest is the victim/suspect racial/ethnic dyad.⁶ The majority of victims were Hispanic, accounting for 50.8% of the sample (n = 333). In addition, 29.9% of the victims were White (n = 196), and 19.2% were Black (n = 126). Of the first identified suspect, 21.4% were White (n = 140), 24.9% were Black (n = 163), and 53.7% were Hispanic (n = 352). The victim/suspect racial/ethnic dyad is measured using the following dummy variables: suspect Black/victim Black (16.3%), suspect Black/victim Hispanic (4.3%), suspect Black/victim White (4.3%), suspect Hispanic/victim Black (2.0%), suspect Hispanic/victim Hispanic (42.0%), suspect Hispanic/victim White (9.8%), suspect White/victim Black (0.9%), and suspect White/victim Hispanic (4.6%), with suspect White/victim White (15.9%) as the reference category.

Control Variables

A number of variables were included in the analysis as statistical controls to isolate the effects of the victim/suspect racial/ethnic dyad on the police decision to arrest. In other words, we hold constant theoretically relevant variables to reduce the effect of confounding variables. Following Spohn and Tellis (2014), control variables are broken into three categories, including victim and suspect characteristics, evidentiary strength, and case factors. Regarding victim and suspect characteristics, continuous variables for victim age and suspect age are included (see Spohn & Tellis, 2012, 2014). Victims in this sample had an average age of 26; suspects had an average age of 32. We include variables for the victim/suspect relationship, which is measured by three dummy variables (intimate, nonstranger, and stranger, with stranger as the reference group). In addition, we include a measure of victim risk-taking behavior that is coded 1 if the victim consumed alcohol before or during the incident. In this sample, 152 victims (23.3%) reported consuming alcohol. A variable is included that taps into whether the investigating officer questioned the victim’s credibility. Complainant credibility was coded dichotomously based on the officer’s perception as indicated in the police report (1 = investigating officer questioned the complainant’s credibility, 0 = investigating officer did not question the complainant’s credibility). Cases were coded 1 if the officer explicitly recorded questioning the complainant’s credibility (e.g., “Victim is not credible”) in the investigative report; no interpretations about credibility or officer perceptions were made by the coders. Investigating officers questioned the victim’s credibility in 39 cases (6.0%). We also include a variable that measures what agency the case was reported to (LAPD = 1, LASD = 0). The majority of cases in this sample were under the jurisdiction of the LASD (64.6%).

Tapping into strength of evidence in a case, the first variable measures whether there was any type of physical evidence collected from the scene of the incident or from the victim or suspect. A composite measure of evidence was included that was coded 1 if any of the following types of evidence were collected: clothing, semen, skin, fingerprints, blood, hair, or bedding. Physical evidence was collected in 41.2% (n = 270) of the cases in this sample. We also include two variables that measure whether the suspect and victim were willing to participate in the investigation, including whether the suspect was interviewed by the police (yes = 1, no = 0), and whether the victim was willing to cooperate in the prosecution of the suspect during the investigation (yes = 1, no = 0). Regarding victim cooperation, investigating officers recorded the victim’s decision to cooperate at three cooperation stages: at the time of initial reporting, once the investigation began (variable used in the current study) and after arrest. Suspects were interviewed in 388 cases (59.2%), and the majority of victims were willing to cooperate at the investigation stage (74.1%, n = 484). A dichotomous measure was also included that indicated whether there were witnesses to the incident (witnesses = 1, no witnesses = 0). This variable was coded 1 if the report indicated the presence of a fresh complaint witness and/or if the victim reported that one or more witnesses were present to the incident. Almost half of the cases in this sample had at least one witness (46.0%, n = 301). Whether the victim made a prompt report was included; this was defined as a report within 1 hr of the incident (1 hr = 1, more than 1 hr = 0). One fourth (n = 162) of the victims in this sample reported the incident to law enforcement within 1 hr.

Regarding case factors, we include whether the incident was an aggravated assault (57.6%), which includes incidents where the suspect used a weapon during the attack (9.6%) and/or whether the victim suffered collateral injury. In this sample, 43.1% (n = 282) of the victims suffered some type of physical injury. The injury variable captures whether the investigating officer recorded victim injuries, whether the sexual assault nurse examiner (SANE) recorded victim injuries during the forensic medical exam, and—in cases where the victim delayed in reporting—whether the victim reported that she suffered injuries during the attack. We also include a variable that measures whether the suspect physically assaulted the victim (yes = 1, no = 0) at the time of the incident. In this sample, approximately half (45.2%, n = 296) of the suspects physically assaulted the victim during the incident. Last, we include a dichotomous variable that measures whether the victim physically and/or verbally resisted the suspect (yes = 1, no = 0). The majority of the victims resisted either physically and/or verbally (75.7%, n = 496).

Analytic Strategy

The analyses in the present study are conducted in two stages. To ensure the absence of mulitcollinearity in our sample, we begin by calculating the variance inflation factors (VIFs), tolerance levels, and condition index scores for the variables included in the model. We then estimate the effects of the victim/suspect racial/ethnic dyad on the police decision to arrest. Because our dependent variable is dichotomized, a logistic regression model will be used to estimate the effects of the victim/suspect racial/ethnic dyad on the police decision to arrest.

Results

Table 2 presents the results from the tests for multicollinearity. We calculated the VIFs, tolerance levels, and the condition index scores from each predictor as an explicit test of multicollinearity (see Pratt & Godsey, 2003). As Table 2 indicates, none of the VIFs exceed the standard “cutoff” point of 4 (Fox, 1991; M VIF = 1.44). In addition, the condition index scores for each predictor fell below the cutoff of 30 (Belsley, Kuh, & Welsch, 1980; condition number = 20.86).

Table 2.

Testing for Multicollinearity.

	VIF	Tolerance
Dependent variable
Police arrested the suspect	1.48	.677
Independent variable
Suspect Black/victim Black	1.86	.538
Suspect Black/victim Hispanic	1.29	.774
Suspect Black/victim White	1.31	.762
Suspect Hispanic/victim black	1.14	.875
Suspect Hispanic/victim Hispanic	2.25	.445
Suspect Hispanic/victim White	1.53	.655
Suspect white/victim Black	1.08	.923
Suspect White/victim Hispanic	1.27	.786
Control variables
Los Angeles Police Department	1.11	.899
Suspect/victim characteristics
Intimate	2.33	.430
Nonstranger	2.61	.383
Victim age	1.44	.696
Suspect age	1.35	.739
Victim consumed alcohol	1.26	.793
Victim’s credibility questioned	1.06	.941
Strength of evidence
Physical evidence collected	1.37	.733
Suspect interviewed	1.21	.825
Victim cooperated	1.24	.805
At least one witness	1.15	.867
Prompt reporting	1.35	.741
Case factors
Aggravated sexual assault	1.57	.638
Suspect physically assaulted victim	1.26	.796
Victim resisted	1.12	.891
M VIF =	1.44
Condition number =	20.86

Note. VIF = variance inflation factors.

The results of the analysis testing the sexual stratification hypothesis are presented in Table 3. As these data indicate, the police were less likely to make an arrest in cases involving Black suspects who assaulted Black victims or Hispanic victims. In addition, police agency, victim characteristics, strength of evidence indicators, and measures of case factors predict the police decision to make an arrest. Regarding victim characteristics, victim alcohol consumption is significant (b = −0.59, p < .05), with the police less likely to make an arrest in cases involving a victim who was drinking before or at the time of the incident. Strength of evidence measures that were found to be significant include whether the suspect was interviewed (b = 1.57, p < .01), whether the victim cooperated at the time of investigation (b = 1.75, p < .01), whether the victim reported the presence of one or more witnesses (b = 0.42, p < .05; fresh complaint and/or eye witness), and whether the victim reported the incident within 1 hr (b = 1.55, p < .01). All of these factors increased the likelihood that the police would make an arrest. In addition, whether physical evidence was collected also influenced whether an arrest was made (b = 0.39, p < .10). Although the conclusions drawn from this relationship should not be overstated as it was only significant at the .10 level. It should also be noted that the police were almost 6 times more likely to arrest the suspect in cases involving a cooperative victim and 5 times more likely in cases where the suspect was interviewed. Case factors that increased the likelihood of arrest include whether the suspect physically assaulted the victim during the sexual attack (b = 0.53, p < .05).

Table 3.

Results of the Logistic Regression Analysis: The Decision to Arrest.

	B	SE	Exp(b)
Independent variable
Suspect Black/victim Black	−0.622^†	0.356	0.537
Suspect Black/victim Hispanic	−1.347*	0.562	0.260
Suspect Black/victim White	−0.436	0.524	0.646
Suspect Hispanic/victim Black	0.626	0.875	1.870
Suspect Hispanic/victim Hispanic	−0.252	0.293	0.777
Suspect Hispanic/victim White	−0.465	0.399	0.628
Suspect white/victim Black	0.849	0.965	2.338
Suspect White/victim Hispanic	0.663	0.532	1.940
Control variables
Los Angeles Police Department	0.419*	0.217	1.521
Suspect/victim characteristics
Intimate	0.160	0.338	1.174
Nonstranger	0.196	0.322	1.216
Victim age	−0.010	0.010	0.990
Suspect age	0.009	0.010	1.009
Victim consumed alcohol	−0.592*	0.258	0.553
Victim’s credibility questioned	−0.627	0.407	0.534
Strength of evidence
Physical evidence collected	0.389^†	0.230	1.475
Suspect interviewed	1.573**	0.204	4.823
Victim cooperated	1.746**	0.252	5.733
At least one witness	0.424*	0.208	1.528
Prompt reporting	1.545**	0.278	4.686
Case factors
Aggravated sexual assault	0.125	0.244	1.134
Suspect physically assaulted victim	0.531*	0.220	1.701
Victim resisted	−0.018	0.241	0.982
Constant	−2.957	0.599	0.052
Nagelkerke R² = .42
Hosmer and Lemeshow Test	Chi-square	df	Sig
	13.122	8	.108

Note. Entries are unstandardized coefficients. SE = standard error.

†

p < .10. *p < .05. **p < .01 (two-tailed test).

Discussion

Building on prior research, this study tested whether the sexual stratification hypothesis can be extended to the decision to arrest using data from one of the largest police departments (LAPD) in the United States. Accordingly, three broad conclusions can be extracted from our results. First, the findings presented here suggest that the police consider the victim/suspect racial/ethnic dyad when making arrest decisions in sexual assault cases. Specifically, the police were less likely to arrest Black suspects who assaulted Black or Hispanic victims than to arrest White suspects who assaulted White victims. Accordingly, our results provide partial support for the sexual stratification hypothesis at the arrest stage. We did not find that Black or Hispanic suspects who assaulted White victims received harsher treatment. Instead, we found a differential response to victimization for women of color. This aligns with the sexual stratification hypothesis’ proposition that maintains that criminal justice action toward victimization will differ depending on the victim/suspect racial/ethnic dyad to the neglect of victims of color (LaFree, 1989; Walsh, 1987). Our findings coincide with previous research that has demonstrated that law-enforcement personnel take crimes against Blacks and Hispanics less seriously than crimes against Whites (see Baldus, Pulaski, & Woodworth, 1983; LaFree, 1989; Walsh, 1987).

Second and relatedly, our findings speak to the body of literature that posits the necessity of looking beyond the main effect of race and ethnicity in trying to detect overt discrimination (see Zatz, 2000 for an overview). Instead, the effects of race and ethnicity need to be further contextualized to discover the more subtle ways implicit bias can result in racial and ethnic disparities (Spohn, 2000). Prior literature has shown that the effects of race on arrest decisions generally can be contextualized according to community characteristics, departmental factors, and individual demographics, such as age and gender (Brown, 2005; Eitle, D’Alessio, & Stolzenberg, 2002; Eitle & Monahan, 2009; Eitle, Stolzenberg, & D’Alessio, 2005; Parker & Maggard, 2005; Stolzenberg, D’Alessio, & Eitle, 2004). Our study demonstrates that the effects of race and ethnicity on the decision to arrest in sexual assault cases specifically can be further contextualized according to victim characteristics where Blacks are treated more leniently given a Black or Hispanic victim. Therefore, not taking the victim/suspect dyad into consideration may mask the effect of race and ethnicity on police decision making in sexual assault cases.

Third, this study found that police officers, in addition to the victim/suspect racial/ethnic dyad, consider legal factors when deciding whether (or not) to arrest a suspect in a case of sexual assault. Whereas extra-legal factors—such as the victim/suspect racial/ethnic dyad and victim alcohol consumption—are legally irrelevant characteristics, legal factors are those that should influence the decision to arrest. Legal factors found to be relevant in this sample include case factors and strength of evidence measures. Specifically, cooperative victims, interviewed suspects, witness presence, prompt reporting, and whether the suspect physically assaulted the victim at the time of the sexual attack all increased the likelihood of arrest. Understanding the factors in general that influence police arrest decisions for sexual assault cases is important because their discretionary powers can greatly affect the outcome of a case.

Returning to the first two points above, findings from this study are particularly important given that research suggests that victims residing in urbanized predominantly Black neighborhoods are sometimes reluctant to cooperate with law enforcement (Miller, 2008). Consistent with the sexual stratification hypothesis, the current study found that arrest was less likely in cases involving women of color who were assaulted by Blacks. Citizens receive powerful messages regarding their social standing through their interactions with legal authorities (Tyler, 1989; Tyler, Degoey, & Smith, 1996). And, the decreased likelihood of arrest involving women of color may perpetuate the historical belief that sexual assault committed against women of color is less severe than the sexual assault of Whites. Fortunately, the police subculture is a powerful entity; it communicates to its members various expectations about police work, ethics, interactions with fellow officers, and attitudes toward victims of color (Adcox, 2000; Rose & Unnithan, 2015). Police officers are positioned to develop and put into action practices that not only increase protection to all sexual assault victims (regardless of race and ethnicity) but also have the potential to increase victim cooperation and decrease case attrition.

The findings of the current study suggest that law-enforcement officers need to actively work toward dismantling beliefs surrounding race/ethnicity and sexual assault, or they run the risk of engaging in policing that denies full protection to certain types of victims. Along the same lines, developing appropriate and supportive response techniques regarding the sexual assault of victims of color is particularly salient to law-enforcement officers who are often the first point of contact victims have with the criminal justice system. Research indicates that only one fourth to one third of sexual assault cases reported to law enforcement will end in an arrest (Alderden & Ullman, 2012a, 2012b; Spohn & Tellis, 2014). Of those cases presented to the prosecutor for filing consideration, fewer than half will result in felony charges (Alderden & Ullman, 2012a). Our study shows that attrition may result from the fact that criminal justice officials support the belief that sexual assault committed against women of color is less severe than the sexual assault of Whites. Therefore, it is imperative that police agencies work at dismantling these beliefs if they want to decrease attrition in these types of cases.

Findings from this study also suggest the need to implement training and education efforts that improve the response to sexual assault victims. This implication is particularly timely given the recent executive order establishing the Task Force on 21st Century Policing. Last year, President Barack Obama established the task force in efforts of providing recommendations regarding effective policing strategies that increase public trust and responsiveness. The task force identified six areas in need of improvement, including officer training and education (President’s Task Force on 21st Century Policing, 2015). Specific to the current study’s findings, the task force suggested that law enforcement agencies work to (a) increase education regarding biases, (b) develop appropriate crisis intervention strategies, and (c) advance victim services. Certainly, this study supports the task force’s call for increased basic and in-service training regarding the response to special populations (i.e., women of color).

Given the implications of the present study, there are numerous possible paths for future research. First, this study contributes to prior work testing the sexual stratification hypothesis; however, there remains substantial opportunity for testing the sexual stratification hypothesis at the arrest stage. Future research should replicate this study to determine whether the same findings cut across different nonurban community types (i.e., rural, suburban), as well as other urban locations that have different law enforcement–community race relations as well as non-U.S. locations. Explanations outside the sexual stratification hypothesis such as issues of victim cooperation, prompt reporting, and presence of witnesses may contribute to further contextualizing the effects of race and ethnicity in sexual assault criminal justice decision making.

It is important to note that the present study relied on sexual assault cases that were reported to two Los Angeles law-enforcement agencies; therefore, it is limited in generalizability. Although studies of sexual assault that rely on police reports are limited because they reflect only those cases that become known to law enforcement, focusing on these cases is justified given that our study examines police decision making in sexual assault cases that are reported to law enforcement (Bergen, 2004). Police reports are an appropriate source of data because they provide detailed information regarding offender, victim, assault, and case-processing characteristics—information that is necessary for the present study. In addition, such data are appropriate for identifying the factors that affect decision making by police. A final limitation of the present study includes our inability to verify the accuracy of the information in each case file. Although we were provided with a redacted copy of each case, we cannot know whether the information provided by the investigating officer accurately represents the victim’s, suspect’s, and witnesses’ experiences.

In the end, this study represents a necessary attempt at understanding the nuances of arrest decisions, a salient issue in the study of criminal justice response to sexual assault. By finding support for the sexual stratification hypothesis at the arrest stage, our study indicates that race and ethnicity factor into police decision making to the detriment of victims of color. Therefore, our study can inform research and policy that aims to create a more equal criminal justice system on additional ways that race and ethnicity can significantly shape sexual assault case outcomes.

Footnotes

Acknowledgements

The authors wish to express their gratitude to the Los Angeles Police Department and the Los Angeles County Sheriff’s Department, both of which provided the redacted case files used for this study.

Authors’ Note

Eryn Nicole O’Neal and Laura O. Beckman contributed equally to this article.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This project was supported by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice [2009-WG-BX-009].

Notes

Author Biographies

Eryn Nicole O’Neal is an assistant professor in the Department of Criminal Justice and Criminology at Sam Houston State University. Her research has appeared in a variety of scholarly journals, including Violence Against Women, Feminist Criminology, Women & Criminal Justice, and Youth Violence and Juvenile Justice. She recently received the American Society of Criminology (ASC) Division on Women and Crime (DWC) Student Paper Competition Award for her research examining victim cooperation in intimate partner sexual assault cases. She also received the ASC DWC 2015 Graduate Scholar Award.

Laura O. Beckman is a PhD candidate and research assistant at Arizona State University. Her research focuses on issues in race, ethnicity, crime and justice.

Cassia Spohn is a foundation professor and the director of the School of Criminology and Criminal Justice at Arizona State University. She is the author or co-author of seven books, including Policing and Prosecuting Sexual Assault: Inside the Criminal Justice System, which was published in 2014. Her research interests include prosecutorial and judicial decision making; the intersections of race, ethnicity, crime and justice; and sexual assault case processing decisions.

References

Adcox

(2000). Doing bad things for good reasons. Police Chief, 67, 16-28.

Alderden

M. A.

Ullman

S. E.

(2012a). Creating a more complete and current picture: Examining police and prosecutor decision-making when processing sexual assault cases. Violence Against Women, 8, 525-551.

Alderden

M. A.

Ullman

S. E.

(2012b). Gender difference or indifference? Detective decision making in sexual assault cases. Journal of Interpersonal Violence, 27, 3-22.

Ancheta

(2006). Race, rights, and the Asian American experience (2nd ed.). Newark, NJ: Rutgers University Press.

Bachman

(1998). The factors related to rape reporting behavior and arrest: New evidence from the National Crime Victimization Survey. Criminal Justice Behavior, 25, 8-29.

Baldus

Pulaski

Woodworth

(1983). Comparative review of death sentences: An empirical study of the Georgia experience. The Journal of Criminal Law and Criminology, 74, 661-753.

Beichner

Spohn

(2005). Prosecutorial charging decisions in sexual assault cases: Examining the impact of a specialized prosecution unit. Criminal Justice Police Review, 16, 461-498.

Belknap

(2007). The invisible woman: Gender, crime, and justice (3rd ed.). Belmont, CA: Wadsworth.

Belsley

D. A.

Kuh

Welsch

R. E.

(1980). Regression diagnostics: Identifying influential data and sources of collinearity. New York, NY: John Wiley.

10.

Bergen

(2004). Wife rape: Reflections on the past, present, and future. Violence Against Women, 10, 1407-1416. doi:10.1177/1077801204270557

11.

Bouffard

J. A.

(2000). Predicting type of sexual assault case closure from victim, suspect, and case characteristics. Journal of Criminal Justice, 28, 527-542.

12.

Brown

R. A.

(2005). Black, white, and unequal: Examining situational determinants of arrest decisions from police-suspect encounters. Criminal Justice Studies, 18, 51-68.

13.

Brownmiller

(1975). Against our will: Men, women and rape. New York, NY: Bantam.

14.

Bryden

D. P.

Lengnick

(1997). Rape in the criminal justice system. The Journal of Criminal Law and Criminology, 87, 1194-1384.

15.

Campbell

(2005). What really happened? A validation study of rape survivors’ help-seeking experiences with the legal and medical systems. Violence and Victims, 20, 55-68.

16.

Campbell

(2006). Rape survivors’ experiences with the legal and medical systems: Do rape victim advocates make a difference? Violence Against Women, 12, 30-45.

17.

Campbell

Patterson

Bybee

Dworkin

E. R

. (2009). Predicting sexual assault prosecution outcomes: The role of medical forensic evidence collected by sexual assault nurse examiners. Criminal Justice and Behavior, 36, 712-727.

18.

Du Mont

Miller

K.-L.

Myhr

J. L

. (2003). The role of “real rape” and “real victim” stereotypes in the police reporting practices of sexually assaulted women. Violence Against Women, 9, 466-486.

19.

Eitle

D’Alessio

S. J.

Stolzenberg

(2002). Racial threat and social control: A test of the political, economic, and threat of black crime hypothesis. Social Forces, 81, 557-576.

20.

Eitle

Monahan

(2009). Revisiting the racial threat thesis: The role of police organizational characteristics in predicting race-specific drug arrest rates. Justice Quarterly, 26, 528-561.

21.

Eitle

Stolzenberg

D’Alessio

S. J.

(2005). Police organizational factors, the racial composition of the police, and the probability of arrest. Justice Quarterly, 22, 30-57.

22.

Felson

R. B.

Pare

P. P.

(2005). The reporting of domestic violence and sexual assault by nonstrangers to the police. Journal of Marriage and Family, 67, 597-610.

23.

Felson

R. B.

Pare

P. P.

(2007). Does the criminal justice system treat domestic violence and sexual assault offenders leniently? Justice Quarterly, 24, 435-459.

24.

Fisher

B. S.

Daigle

L. E.

Cullen

F. T.

Turner

M. G.

(2003). Reporting sexual victimization to the police and others: Results from a national-level study of college women. Criminal Justice and Behavior, 30, 6-38.

25.

Fisher

S. Z.

(1993). “Just the facts, ma’am”: Lying and the omission of exculpatory evidence in police reports. New England Law Review, 28, 1-62.

26.

Fox

(1991). Regression diagnostics. Newbury Park, CA: Sage.

27.

Frazier

P. A.

Haney

(1996). Sexual assault cases in the legal system: Police, prosecutor, and victim perspectives. Law and Human Behavior, 20, 607-628.

28.

Goldstein

(1960). Police discretion not to invoke the criminal process: Low visibility decisions in the administration of justice. The Yale Law Journal, 69, 543-594.

29.

Greenfield

L. A.

(1997). Sex offenses and offenders: An analysis of data on rape and sexual assault. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

30.

Holcomb

J. E.

Williams

M. R.

Demuth

(2004). White female victims and death penalty disparity research. Justice Quarterly, 21, 877-902.

31.

Kennedy

(1997). Race, crime, and the law. New York, NY: Vintage.

32.

Kerstetter

(1990). Gateway to justice: Police and prosecutorial response to sexual assaults against women. Criminology, 81, 267-313.

33.

Kingsnorth

R. F.

Lopez

Wentworth

Cummings

(1998). Adult sexual assault: The role of racial/ethnic composition in prosecution and sentencing. Journal of Criminal Justice, 26, 359-371.

34.

Kingsnorth

R. F.

MacIntosh

R. C.

Berdahl

Blades

Rossi

(2001). Domestic violence: The role of interracial/ethnic dyads in criminal court processing. Journal of Contemporary Criminal Justice, 17, 123-141.

35.

LaFree

(1980). The effect of sexual stratification by race on official reactions to rape. American Sociological Review, 45, 842-854.

36.

LaFree

(1981). Reactions to social problems: Police decisions in sexual assault cases. Social Problems, 28, 582-594.

37.

LaFree

(1989). Rape and criminal justice: The social construction of sexual assault. Belmont, CA: Wadsworth.

38.

Liska

A. E.

(1992). Social threat and social control. Albany: State University of New York.

39.

Maier

S. L.

(2008). “I have heard horrible stories . . .”: Rape victim advocates’ perceptions of the revictimization of rape victims by the police and medical system. Violence Against Women, 14, 786-808.

40.

Mastrofski

S. D.

(2004). Controlling street-level police discretion. The ANNALS of the American Academy of Political and Social Science, 593, 100-118.

41.

Miller

(2008). Getting played: African American girls, urban inequality, and gendered violence. New York: New York University Press.

42.

Parker

K. F.

Maggard

S. R.

(2005). Structural theories and race-specific drug arrests: What structural factors account for the rise in race-specific drug arrests over time? Crime & Delinquency, 51, 521-547.

43.

Patton

T. O.

Snyder-Yuly

(2007). Any four black men will do: Rape, race, and the ultimate scapegoat. Journal of Black Studies, 37, 859-895.

44.

Planty

Langton

Krebs

Berzofsky

Smiley-McDonald

(2013). Female victims of sexual violence, 1994–2010. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

45.

Pratt

Godsey

T. W.

(2003). Social support, inequality, and homicide: A cross-national test of an integrated theoretical model. Criminology, 41, 101-133.

46.

President’s Task Force on 21st Century Policing. (2015). Final report of the President’s Task Force on 21st Century Policing. Washington, DC: Office of Community Oriented Policing Services.

47.

Reiner

(2010). The politics of the police (4th ed.). Oxford, UK: Oxford University Press.

48.

Rennison

C. M.

(2002). Rape and sexual assault: Reporting to police and medical attention, 1992–2000. Washington, DC: U.S. Department of Justice, Bureau of Justice Statistics.

49.

Rose

Unnithan

(2015). In or out of the group? Police subculture and occupational stress. Policing: An International Journal of Police Strategies & Management, 38, 279-294. doi:10.1108/PIJPSM-10-2014-0111

50.

Rowe

(2007). Rendering visible the invisible: Police discretion, professionalism and decision-making. Policing & Society, 17, 279-294.

51.

Shaw

Campbell

(2013). Predicting sexual assault kit submission among adolescent rape cases treated in forensic nurse examiner programs. Journal of Interpersonal Violence, 28, 3400-3417.

52.

Sidanius

Pratto

(2012). Social dominance theory. In Van Lange

P. A.

Kruglanski

A. W.

Tory Higgins

(Eds.), Handbook of theories of social psychology (Vol. 2, pp. 418-438). London, England: Sage.

53.

Sommers

Goldstein

Baskin

(2014). The intersection of victims’ and offenders’ sex and race/ethnicity on prosecutorial decisions for violent crimes. Justice System Journal, 35, 178-204.

54.

Spohn

(1994). Black-on-Black and Black-on-White crime: A comparison of sexual assault case outcomes. In Kedia

P. R.

(Ed.), Black on Black crime: Facing facts, challenging fictions. Bristol, IN: Wyndham Hall Press.

55.

Spohn

(2000). Thirty years of sentencing reform: The quest for a racially neutral sentencing process. In Horney

(Ed.), Policies, processes, and decisions of the criminal justice system (Criminal Justice, Vol. 3, pp. 427-501). Washington, DC: U.S. Department of Justice.

56.

Spohn

Holleran

(2001). Prosecuting sexual assault: A comparison of charging decisions in sexual assault cases involving strangers, acquaintances, and intimate partners. Justice Quarterly, 18, 651-689.

57.

Spohn

Spears

(1996). The effect of offender and victim characteristics on sexual assault case processing decisions. Justice Quarterly, 13, 649-679.

58.

Spohn

Tellis

K. M.

(2012). Policing and prosecuting sexual assault in Los Angeles city and county: A collaborative study in partnership with the Los Angeles Police Department, the Los Angeles County Sheriff’s Department, and the Los Angeles County District Attorney’s Office (NIJ 2009-WG-BX-009). Washington, DC: U.S. Department of Justice.

59.

Spohn

Tellis

K. M.

(2014). Policing and prosecuting sexual assault: Inside the criminal justice system. Boulder, CO: Lynne Rienner.

60.

Stolzenberg

D’Alessio

S. J.

Eitle

(2004). A multilevel test of racial threat theory. Criminology, 42, 673-698.

61.

Tasca

Rodriguez

Spohn

Koss

M. P.

(2012). Police decision making in sexual assault cases: Predictors of suspect identification and arrest. Journal of Interpersonal Violence, 28, 1157-1177.

62.

Tellis

K. M.

Spohn

C. C.

(2008). The sexual stratification hypothesis revisited: Testing assumptions about simple versus aggravated rape. Journal of Criminal Justice, 36, 252-261.

63.

Tjaden

Thoennes

(1998). Prevalence, incidence, and consequences of violence against women: Findings from the National Violence Against Women Survey. Washington, DC: U.S. Department of Justice, National Institute of Justice.

64.

Tyler

T. R.

(1989). The psychology of procedural justice: A test of the group-value model. Journal of Personality and Social Psychology, 57, 830-838. doi:10.1037//0022-3514.57.5.830

65.

Tyler

T. R.

Degoey

Smith

(1996). Understanding why the justice of group procedures matters: A test of the psychological dynamics of the group-value model. Journal of Personality and Social Psychology, 70, 913-930. doi:10.1037//0022-3514.70.5.913

66.

U.S. Census Bureau. (2010). Census of population, P94-171 redistricting data file. Washington, DC: Author.

67.

Walsh

(1987). The sexual stratification hypothesis and sexual assault in light of the changing conceptions of race. Criminology, 25, 153-173.

68.

Wolfgang

Reidel

(1973). Race, judicial discretion, and the death penalty. The ANNALS of the American Academy of Political and Social Science, 407, 119-133.

69.

Zatz

M. S.

(2000). The convergence of race, ethnicity, gender, and class on court decisionmaking: Looking toward the 21st century. In Horney

(Ed.), Policies, processes, and decisions of the criminal justice system (Criminal Justice, Vol. 3, pp. 503-552). Washington, DC: U.S. Department of Justice.