Factors Affecting Sexual Assault Case Processing: Charging Through Sentencing in a Large Southern County

Abstract

Factors affecting prosecutorial decision-making represent one of the most understudied parts of the criminal justice system. Documenting these influences in relation to sexual assault cases is even more rare. The present study analyzed the complete prosecutorial case files of a large, southern district attorney’s office regarding all adult sexual assault cases received over a three-year period. Logistic regression and continuation ratio modeling were used to determine which factors were related to continued progression through the court system, from charging to sentencing. The findings indicate that cases with older or Latino defendants, as well as cases involving injury to the victim, were significantly more likely to be charged. A continuation ratio model of subsequent case outcomes indicated that factors such as DNA evidence, the use of a weapon, and the inclusion of a victim impact statement increased the likelihood of a case progressing to later stages of the system. The influence of criminal history and the amount of prosecutor contact with the victim, however, varied across outcomes. Namely, criminal history increased the odds of receiving a prison sentence while prosecutor contacts with the victim increased the odds of case indictment. These findings imply potential shifts in the treatment of these cases while also suggesting areas of improvement. Namely, prosecutors should strive to increase the amount of meaningful contacts with victims and encourage their participation in the court process. These findings also support the use of sexual assault packets by law enforcement to improve and standardize reporting practices for these cases.

Keywords

sexual assault offenders adult victims

Introduction

Research indicates that sexual assaults are the most underreported violent crime (Morgan & Oudekerk, 2019; National Sexual Violence Resource Center, 2015). According to the National Crime Victimization Survey (NCVS), in 2018 there were an estimated 734,630 sexual assaults or 2.7 sexual assault victimizations per 1,000 persons who were 12 years or older. Approximately 25%, however, were reported to police compared to 63% of robberies and 61% of aggravated assaults (Morgan & Oudekerk, 2019). There are numerous reasons why these crimes are underreported, including victims’ fears of not being believed and their expectations that law enforcement and the court systems will fail to properly handle their cases (see Cantor et al., 2015).

Of those sexual assault cases reported to police, previous research shows that only a small fraction are prosecuted. Research suggests that a prosecutor’s decisions about a case may depend on its conformity to sexual assault stereotypes of “real rapes” and “genuine victims” (Beichner & Spohn, 2012; Campbell et al., 2009). Estrich (1987) referred to real rapes as those that involve a stranger with a weapon who jumps out and sexually assaults a victim. Victims associated with this socially constructed image are more likely to be viewed as “genuine” and their cases are more likely to receive attention through all levels of the criminal justice system (Bryden & Lengnick, 1997; Estrich, 1987). However, the typical sexual assault case does not fit this stereotype.

While more research has been conducted on law enforcement and juror responses to sexual assault cases, the prosecutorial process has been largely inaccessible for research purposes, resulting in a relative dearth of empirical investigation conducted in only a handful of cities. In addition, most existing research relies upon data from the 1990s and early 2000s (e.g., Alderden & Ullman, 2012; Campbell et al., 2009; Frazier & Haney, 1996; Spohn et al., 2001; Spohn & Holleran, 2001) and may not reflect changes in perceptions and handling of sexual assault cases. Finally, research on case decision-making tends to analyze separately run regression analyses predicting each stage of the criminal justice process. This method, however, ignores the conditional nature of each of these stages where attrition occurs at each decision point.

To address these gaps in the literature, with access to the computerized filing system of the district attorney’s (DA) office for a large southern county—a previously unexplored region of the country—this study uses continuation ratio modeling to analyze predictors of adult sexual assault cases moving further through the court system. All sexual assaults, attempted sexual assaults, and aggravated sexual assaults received by the DA’s office from local law enforcement between 2015 and 2017 were examined regarding case outcomes. Specifically, the research question here is which relevant variables predict the charging decision, indictments, and sentence type, and do their impacts vary across outcomes? Based on the research literature, we hypothesize that traditional victim-blaming factors (such as being under the influence or sustaining injury) will predict case outcomes as prosecutors are responsive to potential juror perceptions. Additionally, we expect the prosecutors’ involvement with the victim to show a positive relationship with case progression as more contact should help form a more convictable case with better victim involvement. This work adds to the limited body of recent research in the area and suggests areas of future inquiry.

Literature Review

Prior research indicates that prosecutors’ charging decisions on sexual assault cases are based on the likelihood of receiving a conviction (Alderden & Ullman, 2012; Spohn et al., 2001; Spohn & Holleran, 2001). Studies demonstrate that prosecutors exercise discretion and reject a significant percentage of cases they receive from law enforcement (Campbell et al., 2009; Frazier & Haney, 1996; Spears & Spohn, 1997). Prosecutors have been shown to receive approximately 18%–44% of all sexual assaults reported to law enforcement and pursue charges in 46%–72% of these cases (Bouffard, 2000; Campbell et al., 2009; Campbell et al., 2001; Frazier & Haney, 1996; Spohn et al., 2001; Spohn & Horney, 1993). In total, 14%–18% of these received cases are ever prosecuted (Campbell et al., 2009). This relatively low prosecution rate for sexual assault cases warrants additional study of all points in the process.

Charging Decisions

Campbell et al. (2009) suggested that only 14%–18% of sexual assault cases are ever prosecuted. In a recent study, Morabito et al. (2019) examined complaints of sexual assault in six jurisdictions between 2008 and 2010. Of 3,269 complaints reported to police, charges were filed in 11% of cases; much below the estimate of 75% for felonies in general (Albonetti, 1987). Research examining prosecutors’ sexual assault charging decisions has found that it rests largely on legally relevant factors such as case seriousness, prior record of the offender, and strength of evidence (Albonetti, 1987; Alderden & Ullman, 2012; Beichner & Spohn, 2005; Frazier & Haney, 1996; Kingsnorth et al., 1999; Spohn & Spears, 1996; Spohn & Tellis, 2012; Tellis & Spohn, 2008). Similarly, charging chances have been shown to increase for older victims (see Kingsnorth et al., 1999; Rose & Randall, 1982; Spears & Spohn, 1997; Spohn & Spears, 1996) or when physical evidence, such as victim injury, corroborated the victim’s account of what happened (Henning & Feder, 2005; Spohn & Holleran, 2001).

Factors relevant to risk-taking behavior and the victim–offender relationship have also been shown to be related to charging decisions as well. Victims who used drugs and/or alcohol have been shown to experience a decreased likelihood of their case being pursued (Beichner & Spohn, 2005, 2012; Campbell, 1998; Campbell et al., 2001; Spohn & Holleran, 2001). Using NIBRS data, Richards et al. (2019) found that prosecutors were more likely to decline a case when the offender was an acquaintance, followed by a current or former partner. Spohn and Tellis (2012) argue that cases involving nonstrangers are the most difficult to successfully prosecute. This research suggests that prosecutors believe judges and juries will develop a negative perception of these victims, finding them to be at least partially culpable for or unable to accurately recall their assault. Therefore, they choose to reject the case rather than risk a nonconviction.

Prosecutorial discretion in charging sexual assault cases is particularly complex. Even factors such as the extent to which victims met with prosecutors, the level of victim cooperation, and the victim’s perceived honesty about the offense may influence charging decisions (Spohn et al. (2001). Examinations of the impact of victim and offender race on case processing decisions, though, have generated inconsistent findings. Some research has found that race did have an impact (LaFree, 1989; Spohn & Holleran, 2001), however others have found that victim and offender race and ethnicity were not significantly related to case processing decisions (Kingsnorth et al., 1998; Tellis & Spohn, 2008).

Some of these factors are believed to have an impact because the victim has an inordinate responsibility in these cases to convince a prosecutor, grand jury, and a judge or jury that an offense occurred (O’Neal et al., 2015). Prosecutors are more likely to pursue charges if the victim is deemed credible (Lievore, 2004) and therefore less likely to do so if they question the victim’s credibility (O’Neal et al., 2015). Frohmann (1991, p. 214) suggested that “prosecutors are actively looking for ‘holes’ or problems that will make the victim’s version of ‘what happened’ unbelievable or not convincing beyond a reasonable doubt.”

Case Outcome

The nature of sexual assault often lends itself to circumstantial evidence and contradictory accounts between defendant and complainant (Temken & Krahé, 2008). This creates difficulties for prosecutors, grand juries, juries, and judges, and lowers the likelihood of an indictment and conviction. Studies have found that it is easier to get a conviction when the victim does not know the offender (see Beichner & Spohn, 2012; Loh, 1980; Spohn & Holleran, 2001) and that perpetrators who were related to their victim received less serious charges compared to those with no relationship (Bradmiller & Walters, 1985; McCormick et al., 1998).

Recent research has focused on the fact that less than 1% of all sexual assault cases ever result in a conviction (Anderson, 2019; RAINN, 2020; Van Dam, 2018), which is much lower than rates for other violent crimes (e.g., 4% of assaults result in a conviction; RAINN, 2020). Concerns for the limited response to sexual assault cases received legal attention in 2018, when the city of Austin, Texas faced a civil lawsuit regarding their practices. Only 77 of the 224 cases law enforcement referred to the city’s DA office between 2016 and 2017 were accepted. Of these 77 cases, only 1 went to trial. A total of 10 pleaded to lesser charges, 10 were dismissed, 8 defendants entered a plea as charged, and 7 pleaded to a different felony. Over 30 cases were dismissed and the rest were listed as active when the suit was filed (Amy Smith, Julie Ann Nitsch, & Marina Conner v. City of Austin, 2018).

Much of the research on sexual assault prosecution is decades old (see Campbell et al., 2009 for a review) and involves limited jurisdictions. Newer research shows evidence of lower conviction rates than other crimes but empirical research regarding prosecutorial case processing is limited. Many researchers acknowledge that data regarding prosecutorial decision-making is difficult to obtain (see Bushway & Forst, 2013) and these offices are notoriously unwilling to submit their records to empirical scrutiny. Further research is warranted due to the aged data on the topic from a handful of agencies. The present analyses of prosecutor case files from a large, southern DA’s office examines the extent to which relevant variables predict charging decisions and subsequent case outcomes to add to this important literature.

Methods

The data for this study were derived from the complete prosecutorial case files of the DA’s office of a large southern county and represent all adult sexual assault cases received by the office from local law enforcement agencies. The DA’s office provided access to their case management software, which is the portal used by law enforcement, prosecutors, and defense attorneys to track cases through the court process. Three years of data, 2015 through 2017, involving adult defendants¹

This state allows for certification of juveniles to the adult system. A total of 9 defendants were under the age of 18 at the time of their offense. One was 15, another was 16, and seven were 17 years old.

and victims were coded for various demographic, offense, and case processing information. The definition of sexual assault in this jurisdiction includes penetration of the anus, sexual organ, or mouth of another person without their consent as well as causing another person to penetrate someone else’s (including the actor’s) anus, sexual organ, or mouth without their consent. Also included were attempted sexual assault and aggravated sexual assault, where force or the threat of force (e.g., a weapon or physical intimidation) was used to gain compliance from the victim.

Originally, 422 cases with charges containing the elements of adult sexual assault were identified by the DA’s office as sexual assault cases and provided to the research team. Cases that had not yet reached disposition or that were dismissed due to refiling or conviction in another case were excluded from the analyses, resulting in a final sample size of 342 cases across the three-year time period. Typically, cases that were dismissed due to conviction in another case reflected the defendant being convicted on a nonsexual charge or against another victim in the same incident that appeared in the dataset. Similarly, cases that were dismissed due to refiling typically appeared again under a new case number in the dataset. Therefore, many of these cases were duplicate cases.

Dependent Variables

Charging decision. The charging decision variable is binary (0 = not charged, 1 = charged) and represents the decision of the DA’s office to move forward with a received case. All cases that were sent to the trial court for prosecution were coded as being charged. In most cases, a reason for not pursuing charges was not provided in the records; although, when a reason was stated, it was commonly due to lack of either forensic evidence or proof of nonconsent to the sexual contact.

Case outcomes. The second dependent variable in this study is an ordinal level measure of how far the case made it in the court process once the DA’s office decided to pursue charges (meaning uncharged cases are not included in this variable). This variable was used in the continuation ratio model to analyze predictors of a case moving to the next stage in the process (explained in more detail further). The case outcome variable has three levels: charging (0 = charged); indictment (1 = indicted); and prison sanction (2 = prison sanction). Conviction was not included in this ordinal variable as too few cases (n = 21) were not convicted upon indictment to be informative.

Independent Variables

Defendant demographics. Several defendant demographics were coded from the case files. Race/ethnicity (black, Latino, white, other), gender (female = 0, male = 1), age, and the defendant’s felony criminal arrest history were included as variables. For the purpose of the predictive analyses, the race variable was recoded into a series of three binary variables of black, Latino, and white/other (with white/other excluded from analysis as the reference category). Criminal history is a count of all prior felony offenses contained in the criminal history report and may underreport felony offenses committed in other states or in more distant years.²

Criminal history was square root transformed and age was log transformed to adjust for skew.

Victim factors. Victim race/ethnicity, gender, and age were included as well in the same manner as the defendant demographics. Age was again log transformed to account for positive skew in this variable. Two variables were included to account for victim involvement in the case. Victims in these jurisdictions are given the option to provide victim impact statements (VIS) at the time of reporting their assaults with law enforcement. A binary variable for whether a victim had completed a VIS (0 = no VIS, 1 = VIS) was coded from the case files. The total number of contacts the victim had with the DA’s office were coded as a continuous variable (and later square root transformed to adjust for skew). Contacts were recorded in the notes section of the case management system and the DA’s office stated that these notes were a valid indicator of the number of times the victim and prosecutor were in contact with one another. “Contact” was operationalized as any phone call, in-person meeting, or text exchange between the victim and a member of the prosecutor’s office. A voicemail, unattended meeting, or unreturned text message was not counted as contact for the purposes of this study as this variable is intended to measure the level of involvement of the victim in the case. Finally, whether the victim admitted to being under the influence of alcohol or drugs during the offense was included as a binary variable (0 = not under the influence of drugs or alcohol, 1 = under the influence of drugs or alcohol).

Offense characteristics. Prosecutor notes, offense reports, and sexual assault nurse examination (SANE) reports were read to code for situational characteristics of the offense. This includes use of a weapon, the victim–offender relationship, the presence of DNA evidence, as well as whether the victim was injured during the attack. Injury was any physical injury noted on the SANE report, ranging from bruising/lacerations to internal bleeding/stab wounds. Weapon use (0 = no weapon, 1 = firearm, knife, or blunt object) and victim–offender relationship (0 = nonstranger, 1 = stranger) were included as binary predictors.

Case processing speed. To account for agency practices, the number of days from the occurrence of the offense to the case being received from the law enforcement agency was coded. To reduce the influence of a few outliers and reflect a natural break in the data, this variable was transformed into a binary indicator of whether or not a case was received by the DA’s office within seven days. Roughly 25% of cases were received from law enforcement within one week ( $\bar{X} = 156, s = 418; m e d i a n o f 45 d a y s)$ .

Analytical Procedure

Analyses of these data progressed in two parts. First, descriptive statistics were assessed with regard to the defendant and victim characteristics as well as the commonalities in offense characteristics and case processing. Second, two separate analyses were conducted to predict case progression through the court system. The first analysis used logistic regression to predict whether a case received by the DA’s office resulted in charges being filed with the court. Charging decisions were analyzed separately from the subsequent continuation ratio model since several variables important to later outcomes perfectly predicted the charging decision and would therefore be dropped from the entire analyses were this step included as an outcome. These variables included the use of a weapon and presence of DNA evidence where all but four cases containing such factors were charged. In addition, charges were pursued in all cases including defendant criminal histories and victim impact statements. Therefore, it was pertinent to analyze the charging decision separately on a smaller group of variables and allow the subsequent case outcomes to be predicted with all relevant and available variables. This had the added benefit of allowing the inclusion of a variable only expected to predict charging decisions, whether the case was received within seven days, to be included in this analysis.

The second analysis uses continuation ratio modeling with the logit link to identify which factors predict whether a case moves on to a subsequent phase of the court process.³

We would like to thank an anonymous reviewer for their input on this analysis.

Continuation ratio modeling, also called stage modeling, is used when an ordinal outcome, such as phases of the court process, is dependent upon decision-making processes and each ordinal increase reflects the result of such decisions (Fullerton & Xu, 2016). The phases included here are indictment by a grand jury and incarceration. Therefore, the continuation ratio model predicts (a) a case moving from being charged to being indicted and (b) a case that has been indicted resulting in a prison sentence.

Results

Descriptive Statistics

Case processing. As noted in Figure 1, charges were pursued in 275 of the 342 total cases (80%). Of those cases that resulted in charges, a sizeable percentage of cases (61%) were indicted and a large majority (87%) of indicted cases resulted in some sort of a conviction. These convictions came almost exclusively in the form of a plea bargain as only 11 were adjudicated guilty through a jury trial. Finally, of those cases that were convicted, half were sanctioned with a prison sentence.

Figure 1.

Case processing for all sexual assaults cases, 2015–2017.

Defendant and victim demographics. Table 1 details the defendant and victim demographics of the sexual assault cases in this county from 2015 to 2017. As the table reflects, black and white defendants each comprised 38% of all defendants, while Latinos represented 22% of defendants. Defendants were overwhelmingly male (99%) with an average age of approximately 32 years old. Finally, the average defendant had roughly 1.50 previous felony convictions. With regard to victims, 20% of victims were Latina, while white victims constituted the majority (57%) of victims and were overrepresented relative to their presence in the county, as were black victims (22%). Victims were almost exclusively female (98%) with an average age of approximately 29 years old, and most victims (56%) were not reported to be under the influence of any substances at the time of the offense. A total of 28% of victims completed a VIS with law enforcement and they had, on average, nearly three distinct interactions with the DA’s office.

Offense characteristics. Table 1 also reports characteristics of the offense and case processing. As the table reflects, in most cases (86%) the defendant did not use a weapon. Firearms, knives, and other objects were used as weapons in only 12% of these cases. Concerning the victim–offender relationship, strangers⁴

Stranger was defined in this study as having no interaction or knowledge of each other prior to the assault.

accounted for 16% of all assaults. Approximately one-third (34%) of cases had DNA evidence linking the defendant to the crime. Lastly, in about one quarter of cases (23%), the DA’s office received the case from law enforcement within one week of the offense. The median number of days to receipt was roughly 45 days.

Table 1.

Case Demographics.

Defendant Demographics (n = 334)	ƒ	%
Defendant Demographics (n = 334)	ƒ	%
Race/ethnicity^a
Black	131	38
Latino	76	22
White	130	38
Other	3	1
Male	340	99
Age ( $\bar{X}$ , s)	32.20	11.62
Felony criminal history^b ( $\bar{X}$ , s)	1.48	2.42
Victim Demographics (n = 342)	ƒ	%
Race/ethnicity^a
Black	74	22
Latino	67	20
White	196	57
Other	3	1
Female^c	332	98
Age ( $\bar{X}$ , s)	28.80	12.22
Victim under influence	147	44
Total prosecutor contacts ( $\bar{X}$ , s)	2.78	2.97
VIS completed^d ( $\bar{X}$ , s)	76	28
Offense Characteristics (n = 342)	Number	Percent
Weapon
No weapon	295	86
Weapon	42	12
Victim–offender relationship
Nonstranger	282	82
Stranger	55	16
Victim injured
No	191	57
Yes	144	43
DNA evidence
No	213	66
Yes	110	34
Received w/in 7 days
No	242	77
Yes	71	23

Note. ^aCategories do not equal 100% due to rounding and missing data for race of two defendants and two victims.

^bPercentage calculated out of 270 due to missing data on 71 unfiled cases.

^cGender unknown in 2 cases.

^dPercentage calculated from filed cases.

Predicting Charging Decisions

A logistic regression model was fitted using those predictor variables that did not perfectly predict the charging decision. Also, victim–prosecutor contacts, criminal history checks, and the processing of DNA most commonly occurred after filing, so these variables were excluded to avoid causal order problems. Finally, multicollinearity resulted in victim age and race/ethnicity (which were moderately correlated with the corresponding defendant demographics⁵

See Appendix A for correlations amongst all variables.

) being excluded from the analysis.

Table 2 depicts findings from the logistic regression model that predicted whether or not the DA’s office decided to charge a case. The odds of a case being filed were reduced by 60% when the victim was under the influence of alcohol or drugs during the assault. Cases involving Latino defendants and older defendants exhibited higher odds of being charged as did those where the victim suffered a physical injury. Finally, the DA’s office was at over five times the odds of filing charges if the case was received within a week of the offense.

Table 2.

Logistic Regression of Charging Decision on Predictor Variables.

	Odds Ratio	b (se(b))	95% Confidence Interval
Defendant age^a	2.75*	1.01 (0.49)	0.04–1.98
Defendant black^b	1.30	0.26 (0.38)	−0.49–1.01
Defendant Latino^b	4.96*	1.60 (0.66)	0.31–2.89
Injury	3.17**	1.15 (0.39)	0.40–1.91
Victim under the influence	0.39**	−0.94 (0.37)	−1.67–−0.20
Stranger	1.43	0.36 (0.53)	−0.68–1.40
Received within 7 days	5.61*	1.72 (0.78)	0.20–3.25
	n	305
	Wald $χ^{2}$	25.31***
	R²	0.17

Note. 37 cases dropped due to missing data on one or more predictor variables. Robust standard errors were estimated.

^aThis variable was log transformed.

^bWhite/other race excluded as the reference category.

*p < .05. **p < .01. ***p < .001.

Continuation Ratio Model of Case Outcomes

The remaining 275 charged cases were then analyzed for case outcomes using continuation ratio modeling. Those variables that were excluded in the previous analysis (VIS, total prosecutor contacts, DNA evidence, and weapon use) were included in the continuation ratio model. This method allows for the prediction of indictments from those cases that were pursued and the prediction of a prison sentence from those that were indicted. In continuation ratio modeling, variables can be permitted to differentially impact the ordinal outcomes (referred to as being unconstrained; i.e., predicting indictment in a different direction than sentencing) while others can be constrained (meaning they are not allowed to freely vary across values of the ordinal outcome). The Brant test, used to assess whether any variables in a model are better included as unconstrained, indicated that at least one variable violated the parallel regression assumption ( $χ^{2}$ = 31.60, p < .001; Hosmer et al., 2013). Both criminal history and prosecutor contact were specifically shown to violate this assumption and to be better included as unconstrained variables (respectively, $χ^{2}$ = 4.02, p < .05; $χ^{2}$ = 15.56, p < .001). Therefore, the result was an unconstrained partial continuation ratio model where the impacts of criminal history and prosecutor contact were allowed to vary across levels of the ordinal outcome while all other variables were assumed to impact each level in the same manner.

Table 3 presents the results of the unconstrained partial continuation ratio model predicting a case progressing further through the court system. Latino defendants experienced over two and a half times the odds of progressing further through the system compared to white defendants. Further, a case had 2.31 times the odds of reaching the next stage if the defendant used a weapon during the assault and the odds increased by almost four times if there was DNA evidence linking the defendant to the crime. The odds were reduced by over 50% if the victim was reported to be under the influence of drugs or alcohol at the time of the assault. Meanwhile, the odds of progressing further through the court process were two times higher if the victim provided an impact statement.

Table 3.

Continuation Ratio Regression of Case Outcomes on Predictor Variables.

	Odds Ratio	b (se(b))	95% Confidence Interval
Constrained Variables
Defendant age^a	0.87	−0.14 (0.35)	−0.83–0.55
Defendant Black^b	1.68	0.52 (0.28)	−0.03–1.06
Defendant Latino^b	2.61**	0.96 (0.30)	0.37–1.55
Weapon	2.31**	0.84 (0.33)	0.20–1.48
Injury	0.80	−0.23 (0.24)	−0.70–0.25
DNA evidence	3.76***	1.33 (0.27)	0.80–1.85
Victim under the influence	0.47**	−0.75 (0.25)	−1.24–−0.27
Stranger	0.88	−0.13 (0.28)	−0.68–0.43
Victim impact statement	2.03**	0.71 (0.26)	0.20–1.21
Unconstrained variables
Criminal history^c
Indictment	1.27	0.24 (0.16)	−0.07–0.55
Prison sentence	1.80**	0.59 (0.21)	0.18–0.99
Prosecutor contact w/ victim^c
Indictment	2.39***	0.87 (0.19)	0.50–1.24
Prison sentence	0.72	−0.33 (0.22)	−0.76–0.11
n	259
	Wald $χ^{2}$	50.28***

Note. 16 cases dropped due to missing data on one or more predictor variables. Robust standard errors were estimated.

^aThis variable was log transformed.

^bWhite/other race excluded as the reference category.

^cThe square root was used to account for positive skew in this variable.

*p < .05. **p < .01. ***p < .001.

Criminal history and the amount of contact the prosecutors had with the victims were allowed to freely vary across levels of the dependent variable. Therefore, odds ratios are reported for these two variables with regard to the odds of indictment and of receiving a prison sentence. Criminal history appeared to be unrelated to indictment, but defendants with more extensive criminal histories were shown to be at increased odds of receiving a prison sentence. With regard to prosecutor contacts, the more contact a prosecutor had with the victim, the higher the odds of indictment. These contacts were not predictive, however, of a defendant being incarcerated.

Discussion

The descriptive data provided many interesting findings that help place sexual assault case processing into context. For example, the finding that the DA’s office decided to charge most sexual assault cases concurs with previous research (Frazier & Haney, 1996; Morabito et al., 2019; Spohn & Holleran, 2001; Spohn & Tellis, 2014). The current study also noted that virtually all indicted cases resulted in some sort of conviction; however, the 146 convictions constitute less than half of all cases brought to the DA’s office. Further, half of convicted cases resulted in a prison sentence. These findings generate interest in determining which factors contribute to a case progressing to later stages in the court process and which factors determine sentencing type.

The inferential results also align with our hypothesis that some indicators of being a “genuine” victim are still associated with successful prosecution. Sober victims and those who sustained injury were more likely to have their cases pursued, indicating that prosecutors are still sensitive to a jury’s adherence to rape myths; perceiving a victim who had used alcohol or drugs as inviting the assault or one who did not have injuries as not fighting back. On the other hand, the victim–offender relationship failed to differentiate the outcomes measured here as it had in previous research (Bradmiller & Walters, 1985; Campbell et al., 2009; Holleran et al., 2010; McCormick et al., 1998; Spohn et al., 2001; Spohn & Tellis, 2014).

Further, the results regarding victim involvement align with the qualitative work of Spohn et al. (2001) who showed that prosecutors consider victim involvement to be an important asset in sexual assault prosecution. The importance of keeping victims involved in the processing of sexual assault cases was also highlighted by Morabito et al. (2019) who found that the involvement of cooperative victims was the strongest predictor of arrest in all of the jurisdictions they studied. The data for the present study are unique in their inclusion of prosecutor notes detailing their contacts with victims. Without yet delving into the qualitative nature of these notes, the sheer number of conversations held between the prosecutors and victims appeared to play a meaningful role in case outcomes. Specifically, victims averaged just under three (2.78) contacts with prosecutors and indictment was more likely with increasing contacts. Another measure of victim involvement, completion of a VIS, also increased the odds of case progression. While victim participation may be hindered by factors outreach efforts seem here to be worth the investment.

In line with past studies, these analyses also found that the presence of physical evidence is significantly related to successful prosecution. Of course, convictions are more easily secured in cases with tangible evidence. The ability to introduce DNA evidence is important in helping to ensure a conviction and avoiding wrongful convictions. Despite much recent attention focused on this type of evidence, little is known about how it influences police and prosecutors’ handling of sexual assault cases in particular (Menaker et al., 2017). It seems reasonable that in light of the CSI effect, prosecutors are more reluctant to move forward with cases that lack this evidence in anticipation of jurors’ already skeptical responses to sexual assault cases. However, forensic evidence is not present in a sizeable number of sexual assault cases (66% in the present study) and even when it is, Peterson et al. (2010) report that the material is often not submitted for analysis. Therefore, the exact role that DNA and other forensic evidence play in prosecutorial decision-making in sexual assault cases is not well-documented, although the present study suggests their continued preeminence.

This study also adds to the conflicting literature on the influence of race on prosecutorial outcomes in these cases. Our findings align with those who have shown race and ethnicity matter in predicting charging decisions and other case outcomes. Many factors likely combine to explain the increased likelihood of more serious outcomes in sexual assault cases involving Latino defendants. Reporting practices, prosecutorial and jury discretion, differences in the quality of defense representation, cultural differences regarding trust in the system, and a host of other factors could each help explain the disparity. Latinas have been shown to be violently victimized at rates exceeding that of their white counterparts (Rennison, 2002), but they are also less likely to report this victimization; often due to fear of deportation or a lack of trust in law enforcement (Messing et al., 2015). Emerging evidence from cities such as Houston, Texas indicates that sexual assault reporting by Latina victims may be decreasing due to the current climate surrounding Latino immigration. The Houston Police Department noted that reports of sexual assault from Latina victims have decreased by 43% despite an increase in reporting of nearly 12% by non-Latina victims (Houston Public Media Staff, 2017); a trend noted in other cities as well (Feeney, 2017; Theodore, 2013). The fact that we find Latino defendants to be more likely to see their cases charged, indicted, and sanctioned with a prison sentence may indicate in part that the cases that were reported were simply more serious cases that involved significant physical injury (or threat of injury) to outweigh the perceived risks of reporting. The present study did not measure injury seriousness to be able to assess this possibility, but it remains an important inquiry for future research.

Another salient factor explaining why Latino, but not black, defendants’ cases progressed further than white defendants’ cases could also stem from the focus on racism against black defendants in this county and across the country without a similar concern for bias against other historically disenfranchised groups. The findings here suggest the need for agencies to move beyond the black–white binary and assess their racially and ethnically disparate outcomes across a wider range of identities. It is important that these agencies work to identify and meaningfully address the sources of racial disparity in their prosecutorial decision-making, the grand jury process, and outreach efforts as well.

Limitations

Findings from the current study must be placed in the context of certain limitations. First, these data stem from a single DA’s office in one southern county of the United States over a three-year time period. The generalizability of the results presented here to other agencies and jurisdictions need to be considered with this in mind. Further, this study relied upon official records and may not reflect the circumstances regarding those cases that are not reported to police or the cases that law enforcement chose to not submit to the DA’s office for prosecution. In fact, this office is attempting to merge their cases and those reported to law enforcement to better understand the interplay between these agencies, which is an important area for future research.

Still, these data represent a quite complete account of those cases that were submitted to this DA’s office. These files contain all police and investigator reports, SANE reports, prosecutor and victim services notes, physical evidence, testimony, as well as criminal background and pretrial detention information. Some variables, however, were still not easily coded. It is important to note that the substance use variable, in particular, may lack reliability. Many police reports in the case files did not mention substance use by either party. It often appeared that it was left to the victim to disclose this information. Additionally, race and ethnicity were often not clearly indicated in the files. Many police report forms did not appear to provide good guidance on this construct so that many Latino defendants were recorded as white. This is an unfortunate manifestation of the black–white binary the system tends to fall into and these reports (and others) should make better attempts to document the complex reality of racial and ethnic identity. In light of this limitation, the researchers used other reports and documents (such as SANE reports) contained within these files to increase the validity of this measure. Finally, it is likely that there were important interactions taking place amongst these variables—including race/ethnicity, injury, substance use, and victim involvement as examples—that factored into these prosecutors’ decisions yet could not be assessed here due to the sample size.

Policy Implications and Directions for Further Study

The present study answered important questions regarding legal responses to sexual assault and highlighted the need for additional work in the area. Findings from the present study identified areas where improvements can be made to ensure that sexual assault cases are properly addressed and generated the need to more closely examine certain areas. These suggestions primarily pertain to changes that could occur in our justice systems, with the goal of improving current practices.

Prosecutors heavily rely on the information brought to their attention via police reports. As such, there is a need to ensure that all evidence is collected and that police reports thoroughly capture the details of each incident. Police investigations and reports should be thorough, and there should be communication and interaction among law enforcement and prosecutors regarding expectations and practices. These suggestions are commonsensical, yet these practices vary by police department. As the entry point to the criminal justice system, police personnel must be aware of the significance of sexual assault cases and the importance of conducting thorough investigations and reports. Qualitatively analyzing the completeness of police reports for this type of crime and the impacts thoroughness has on case outcomes represents an important area for future research.

This and other studies suggest that more victim involvement in the process would enhance the likelihood of cases moving forward. Considerable efforts should be made, if they are not already, to further involve victims in the process. The number of prosecutor contacts was positively related to indictments in sexual assault cases in this county. While not empirically analyzed here, it was apparent from reviewing the prosecutor and victim advocate case notes that numerous victims expressed frustration with their lack of knowledge concerning their case and that responding to these concerns in a meaningful way appeared to increase the victim’s trust and involvement in the process. To be sure, there are sometimes challenges involved with contacting victims (e.g., they may not wish to be involved with the case and actively avoid such contact) and identifying the means by which these obstacles can be overcome requires further investigation.

The creation of a uniform sexual violence packet, akin to those used in domestic violence cases, would provide much helpful information regarding the context and circumstances surrounding these cases. Information is key in today’s society, for instance as police departments are increasingly basing their practices on data analyses (e.g., Brayne, 2017). Compiling much needed information in a single report for each sexual assault case could provide for convenient analyses of trends, patterns, practices, strengths, and limitations with regard to these cases. Beyond the empirical usefulness of such a tool, these packets could serve as vital, brief resources to assist prosecutors in preparing sexual assault cases and allow more immediate analyses to regularly monitor agency practices.

Conclusion

Overall, the results presented here shed more light on the prosecution of sexual assault cases and the outcomes associated with those efforts. Consideration of the results from a large, southern DA’s office provides helpful evidence of how sexual assault cases are currently addressed in a previously unexplored region of the country. Further examination of sexual assault cases is necessary given the seriousness of these offenses, their hidden nature, the historically limited justice-based responses to this crime, and the vast misconceptions surrounding it. The present study helped address some of the concerns regarding the processing of sexual assault cases, but many areas remain in need of additional study. Justice-based practices have undergone much reform through the years, and it can be argued that today’s practices are far better than those of years past. Despite the improvements, much more can be done.

Supplemental Material

Supplemental material for this article available online.

Supplemental Material for Factors Affecting Sexual Assault Case Processing: Charging Through Sentencing in a Large Southern County by Brie Diamond, Kendra Bowen, Ronald Burns, in Journal of Interpersonal Violence

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.

ORCID iD

Brie Diamond

Author Biographies

Brie Diamond, PhD, is an associate professor of criminal justice at Texas Christian University. Her research interests include criminological theory, program evaluation, and criminal case processing. Her work can be found in Journal of Criminal Justice, Intelligence, and Criminal Justice and Behavior.

Kendra Bowen, PhD, is an associate professor of criminal justice at Texas Christian University. Her current research primarily focuses on sex crimes. Specifically, sexual assaults, crimes against children, and various aspects of the sexual abuse scandal in the Catholic Church. Recent publications have appeared in Corrections: Policy, Practice and Research, Sexual Abuse, Journal of Crime, Justice and Social Democracy,and the Journal of Offender Rehabilitation.

Ronald Burns, PhD, is department chair and professor of criminal justice at Texas Christian University. He has published over 75 articles in areas including policing, multiculturalism in the criminal justice system, the criminal justice system, and white-collar crime. He is the author, coauthor, or editor of eight books, including Multiculturalism, Crime and Criminal Justice; Environmental Law, Crime and Justice; Federal Law Enforcement; Policing: A Modular Approach; Critical Issues in Criminal Justice; The Criminal Justice System; Environmental Crime: A Sourcebook, and Policing and Violence. Dr Burns graduated from Florida State University in 1997 and has been at Texas Christian University ever since.

References

Albonetti

(1987). Prosecutorial discretion: The effects of uncertainty. Law & Society Review , 21(2), 291–313. https://doi.org/10.2307/3053523

Alderden

M. A.

, & Ullman

S. E.

(2012). Creating a more complete and current picture: Examining police and prosecutor decision-making when processing sexual assault cases. Violence Against Women , 18(5), 525–551. https://doi.org/10.1177/1077801212453867

Smith

Amy

, Nitsch

Julie Ann

, & v.

Marina Conner

City of Austin. U.S. District Court Western District of Texas. Case No. 1:18-cv-505 (2018). https://www.courthousenews.com/wp-content/uploads/2018/06/SexAssaultTX.pdf

Anderson

(2019, September 15). Why are we so bad at prosecuting sexual assault? The Dallas morning news . https://www.dallasnews.com/opinion/commentary/2019/09/15/why-are-we-so-bad-at-prosecuting-sexual-assault/

Beichner

, & Spohn

(2005). Prosecutorial charging decisions in sexual assault cases: Examining the impact of a specialized prosecution unit. Criminal Justice Police Review , 16(4), 461–498. https://doi.org/10.1177/0887403405277195

Beichner

, & Spohn

(2012). Modeling the effects of victim behavior and moral character on prosecutors’ charging decisions in sexual assault cases. Violence and Victims , 27(1), 3–24. https://doi.org/10.1891/0886-6708.27.1.3

Bouffard

(2000). Predicting type of sexual assault case closure from victim, suspect and case characteristics. Journal of Criminal Justice , 28(6), 527–542. https://doi.org/10.1016/S0047-2352(00)00068-4

Bradmiller

L. L.

, & Walters

W. S.

(1985). Seriousness of sexual assault charges: Influencing factors. Criminal Justice and Behavior , 12(4), 463–484. https://doi.org/10.1177/0093854885012004005

Brayne

(2017). Big data surveillance: The case of policing. American Sociological Review , 82(5), 977–1008. https://doi.org/10.1177/0003122417725865

10.

Bryden

D. P.

, & Lengnick

(1997). Rape in the criminal justice system. Journal of Criminal Law and Criminology , 87(4), 1194–1384. https://doi.org/10.2307/1144018

11.

Bushway

S. D.

, & Forst

(2013). Studying discretion in the processes that generate criminal justice sanctions. Justice Quarterly , 30(2), 199–222. https://doi.org/10.1080/07418825.2012.682604

12.

Campbell

(1998). The community response to rape: Victims’ experiences with the legal, medical, and mental health systems. American Journal of Community Psychology , 26(3), 355–379. https://doi.org/10.1023/a:1022155003633

13.

Campbell

, Patterson

, Bybee

, & Dworkin

E. R.

(2009). Predicting sexual assault prosecution outcomes. Criminal Justice and Behavior , 36(7), 712–727. https://doi.org/10.1177/0093854809335054

14.

Campbell

, Wasco

S. M.

, Ahrens

C. E.

, Sefl

, & Barnes

H. E.

(2001). Preventing the “second rape”: Rape survivors’ experiences with community service providers. Journal of Interpersonal Violence , 16(12), 1239–1259. https://doi.org/10.1177/088626001016012002

15.

Cantor

, Fisher

, Chibnall

, Townsend

, Lee

, Bruce

, & Thomas

(2015). Report on the AAU campus climate survey on sexual assault and sexual misconduct . The Association of American Universities. https://www.aau.edu/sites/default/files/%40%20Files/Climate%20Survey/AAU_Campus_Climate_Survey_12_14_15.pdf

16.

Estrich

(1987). Real rape . Harvard University Press.

17.

Feeney

(2017). May 25). Cities notice decline in Latino crime reporting post-Trump. Cato Institute. https://www.cato.org/blog/cities-notice-decline-latino-crime-reporting-post-trump

18.

Frazier

P. A.

, & Haney

(1996). Sexual assault cases in the legal system: Police, prosecutor, and victim perspectives. Law and Human Behavior , 20(6), 607–628. https://doi.org/10.1007/BF01499234

19.

Frohmann

(1991). Discrediting victims’ allegations of sexual assault: Prosecutorial accounts of case rejections. Social Problems , 38(2), 213–226. https://doi.org/10.2307/800530

20.

Fullerton

A. S.

, & Xu

(2016). Ordered regression models: Parallel, partial, and non-parallel alternatives . Chapman & Hall.

21.

Henning

, & Feder

(2005). Criminal prosecution of domestic violence offenses: An investigation of factors predictive of court outcomes. Criminal Justice and Behavior , 32(6), 612–642. https://doi.org/10.1177/0093854805279945

22.

Holleran

, Beichner

, & Spohn

(2010). Examining charging agreement between police and prosecutors in rape cases. Crime & Delinquency , 56(3), 385–413. https://doi.org/10.1177/0011128707308977

23.

Hosmer

D. W.

, Lemeshow

, & Sturdivant

R. X.

(2013). Applied logistic regression (3rd ed.) . John Wiley & Sons.

24.

Houston Public Media Staff. (2017). Is “sanctuary cities” talk really having a chilling effect in Houston ? https://www.houstonpublicmedia.org/articles/news/2017/05/01/198284/is-sanctuary-city-talk-really-having-a-chilling-effect-in-houston/

25.

Kingsnorth

, Lopez

, Wentworth

, & Cummings

(1998). Adult sexual assault: The role of racial/ethnic composition in prosecuting and sentencing. Journal of Criminal Justice , 26(5), 359–371. https://doi.org/10.1016/S0047-2352(98)00012-9

26.

Kingsnorth

, MacIntosh

, & Wentworth

(1999). Sexual assault: The role of prior relationship and victim characteristics in case processing. Justice Quarterly , 16(2), 275–302. https://doi.org/10.1080/07418829900094141

27.

LaFree

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

28.

Lievore

(2004). Victim credibility in adult sexual assault cases, Trends & Issues in Crime and Criminal Justice, 288, 2–6.

29.

Loh

W. D.

(1980). The impact of common law and reform rape statutes on prosecution: An empirical study. Washington Law Review , 55, 543–562.

30.

McCormick

J. S.

, Maric

, Seto

M. C.

, & Barbaree

H. E.

(1998). Relationship to victim predicts sentence length in sexual assault cases. Journal of Interpersonal Violence , 13(3), 413–420. https://doi.org/10.1177/088626098013003007

31.

Menaker

T. A.

, Campbell

B. A.

, & Wells

(2017). The use of forensic evidence in sexual assault investigations: Perceptions of sex crimes investigators. Violence Against Women , 23(4), 399–425. https://doi.org/10.1177/1077801216641519

32.

Messing

J. T.

, Becerra

, Ward-Lasher

, & Androff

D. K.

(2015). Latinas’ perceptions of law enforcement: Fear of deportation, crime reporting, and trust in the system. Journal of Women and Social Work , 30(3), 328–340. https://doi.org/10.1177/0886109915576520

33.

Morabito

M. S.

, Williams

L. M.

, & Pattavina

(2019). Decision making in sexual assault cases: Replication research on sexual violence case attrition in the U.S. office of justice programs . U.S. Department of Justice. https://www.ncjrs.gov/pdffiles1/nij/grants/252689.pdf

34.

Morgan

R. E.

, & Oudekerk

B. A.

(2019). Criminal victimization, 2018 . U.S. Department of Justice. https://www.bjs.gov/content/pub/pdf/cv18.pdf

35.

National Sexual Violence Resource Center. (2015). Statistics about sexual violence . Author. https://www.nsvrc.org/sites/default/files/publications_nsvrc_factsheet_media-packet_statistics-about-sexual-violence_0.pdf

36.

O’Neal

E. N.

, Tellis

, & Spohn

(2015). Prosecuting intimate partner sexual assault: Legal and extra-legal factors that influence charging decisions. Violence Against Women , 21(10), 1237–1258. https://doi.org/10.1177/1077801215591630

37.

Peterson

J. L.

Sommers

Baskin

Johnson

(2010). The role and impact of forensic evidence in the criminal justice process . U.S. Department of Justice, National Institute of Justice. https://www.ojp.gov/pdffiles1/nij/grants/231977.pdf

38.

RAINN. (2020). The criminal justice system: Statistics. https://www.rainn.org/statistics/criminal-justice-system

39.

Rennison

C. M.

(2002). Hispanic victims of violent crime, 1993-2000 . Bureau of Justice Statistics.

40.

Richards

T. N.

, Tillyer

M. S.

, & Wright

E. M.

(2019). When victims refuse and prosecutors decline: Examining exceptional clearance in sexual assault cases. Crime & Delinquency , 65(4), 474–498. https://doi.org/10.1177/0011128719828351.

41.

Rose

, & Randall

(1982). The impact of investigator perceptions of victim legitimacy on the processing of rape/sexual assault cases. Symbolic Interaction , 5(1), 23–36. https://doi.org/10.1525/si.1982.5.1.23

42.

Spears

, & Spohn

(1997). Prosecutors’ charging decisions in sexual assault cases. Justice Quarterly , 14(3), 501–524. https://doi.org/10.1080/07418829700093451

43.

Spohn

, & Holleran

(2001). Prosecuting sexual assault: A comparison of charging decisions in sexual assault cases involving strangers, acquaintances, and intimate partners. Justice Quarterly , 18(3), 651–688. https://doi.org/10.1080/07418820100095051

44.

Spohn

, & Horney

(1993). Rape law reform and the effect of victim characteristics on case processing. Journal of Quantitative Criminology , 9(4), 383–409. https://www-jstor-org-443.web.bisu.edu.cn/stable/23366015

45.

Spohn

, & Spears

(1996). The effect of offender and victim characteristics on sexual assault case processing decisions. Justice Quarterly , 13(4), 649–679. https://doi.org/10.1080/07418829600093141

46.

Spohn

, & Tellis

(2012). The criminal justice system’s response to sexual violence. Violence Against Women , 18(2), 169–192. https://doi.org/10.1177/1077801212440020

47.

Spohn

, & Tellis

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

48.

Spohn

, Beichner

, & Davis-Frenzel

(2001). Prosecutorial justifications for sexual assault case rejection: Guarding the “gateway to justice.” Social Problems , 48(2), 206–235. https://doi.org/10.1525/sp.2001.48.2.206

49.

Tellis

K. M.

, & Spohn

(2008). The sexual stratification hypothesis revisited: Testing assumptions about simple versus aggravate rape. Journal of Criminal Justice , 36(3), 252–261. https://doi.org/10.1016/j.jcrimjus.2008.04.006

50.

Temken

, & Krahé

(2008). Sexual assault and the justice gap: A question of attitude . Hart Publishing.

51.

Theodore

(2013). Insecure communities: Latino perceptions of police involvement in immigration enforcement . University of Illinois. https://www.policylink.org/sites/default/files/INSECURE_COMMUNITIES_REPORT_FINAL.PDF

52.

Van Dam

(2018 October 6). Less than 1% of rapes lead to felony convictions. At least 89% of victims face emotional and physical consequences. The Washington Post . https://www.washingtonpost.com/business/2018/10/06/less-than-percent-rapes-lead-felony-convictions-least-percent-victims-face-emotional-physical-consequences/

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.31 MB