“That is Not Behavior Consistent With a Rape Victim”: The Effects of Officer Displays of Doubt on Sexual Assault Case Processing and Victim Participation

Abstract

Despite the prevalence and severity of sexual violence, case attrition has been identified as a significant issue. Of the cases that are reported to police, only a small portion result in arrest, prosecution, or conviction. Research has revealed that much of this attrition occurs early in the process and that a number of theoretically supported legal (e.g., physical evidence, victim participation) and extralegal (e.g., demographics, victim credibility) factors influence how and whether a case progresses through the criminal justice system. However, few researchers have directly examined the impact of officer doubt on case processing. Whereas legal and extralegal factors represent case characteristics, doubt represents officer cognition about these characteristics. These perceptions can affect how victims are treated, how police investigate the case, and ultimately, the case’s progression through the system. A random sample of sexual assault reports from one police department in a medium-sized jurisdiction in the western U.S. was drawn to examine the expression of officer doubt, as well as its impact on victim participation, arrest, and referral for prosecution while controlling for relevant legal and extralegal factors. The findings suggest that officer doubt is an important consideration in sexual assault case processing, independent of other legal and extralegal factors, and that it significantly impacts the likelihood of arrest and referral for prosecution. Consistent with previous research, police decision-making was also impacted by certain legal factors. Victim participation was not directly affected by officer doubt but it was predicted by extralegal factors. Implications for future sexual assault research and practitioner training are discussed.

Keywords

sexual assault reporting/disclosure support seeking

Sexual violence is recognized as a public health issue that can result in a number of consequences such as physical injury, posttraumatic stress disorder, and suicidal ideation (Centers for Disease Control [CDC], 2022). In the United States, it is estimated that more than 50% of women and almost one-third of men have been victims of sexual violence involving physical contact in their lifetime (CDC, 2022). The most recent results of the National Crime Victimization Survey indicate that there were almost 325,000 incidents of rape/sexual assault in the United States in 2021 alone (Bureau of Justice Statistics [BJS], 2022). Despite the severity and prevalence of sexual violence, less than 22% of rapes/sexual assaults were reported to police in 2021 (BJS, 2022). There are a variety of reasons why victims choose not to report such as concerns that their credibility will be questioned, the belief that the police will not do anything, mistrust of service providers due to the negative treatment of sexual assault survivors, and fear of reprisal (Alderden & Long, 2016; BJS, 2012; Patterson et al., 2009).

Of the incidents of sexual violence that are reported to police, case attrition is a significant issue (Alderden & Ullman, 2012; Pattavina et al., 2021; Spohn & Tellis, 2019). That is, a small portion of cases lead to arrest, referral for prosecution, charging, and conviction. Research has identified a number of factors that influence case attrition. The presence of physical evidence, crime seriousness, and victim cooperation are examples of legal factors that have been found to affect case progression (Lapsey et al., 2022, 2023; Morabito et al., 2019; Spohn & Tellis, 2019). Extralegal factors, such as victim credibility and victim–suspect relationship, have also been found to predict how a case progresses (Acquaviva et al., 2022; Alderden & Ullman, 2012; Meeker et al., 2021). The effects of these extralegal factors often result from stereotypical beliefs about sexual violence held by criminal justice actors and the public (Pattavina et al., 2021; Sleath & Bull, 2017; St. George et al., 2022). Elements such as a victim’s sexual history and behavior before or after the assault may lead to doubt about whether the victim is being truthful or somehow invited the assault. This doubt can affect how a police officer interacts with a victim and investigates the case, and consequently, whether the victim chooses to participate in the investigation (Alderden & Long, 2016; Growette Bostaph et al., 2021; Morabito et al., 2019).

While several researchers have investigated the effects of legal and extralegal factors on sexual assault case processing, the present study focuses on the impact of officer doubt. Examining only the presence of legal and extralegal factors fails to describe how and why these variables impact case processing. Legal and extralegal factors are case characteristics, whereas doubt represents officers’ cognition about these characteristics. We hypothesize that cases with evidence of officer doubt will be less likely to result in victim participation, arrest, and referral for prosecution while controlling for relevant legal and extralegal factors. A random sample of sexual offense police reports from one medium-sized city in the western U.S. was examined to determine how officer doubt affected case progression.

Literature Review

Sexual assault case attrition has been found to occur at all levels of the criminal justice system, but much of it occurs early in the process when decision-making rests with the police (Alderden & Long, 2016; Lapsey et al., 2022; Sleath & Bull, 2017). Labeled as gatekeepers of the criminal justice system, the police determine how to investigate reports, and often in conjunction with the prosecution, whether to make an arrest (Acquaviva et al., 2022; Morabito et al., 2019; Pattavina et al., 2021; Spohn & Tellis, 2019). Thus, the actions of the police have a demonstrable effect on if and how a sexual assault report progresses through the system.

Arrest and Referral for Prosecution

Research frequently finds that a small portion of sexual assault cases result in arrest (Lapsey et al., 2022, 2023; Morabito et al., 2019; O’Neal & Spohn, 2017). For example, Pattavina et al. (2021) reported that less than one-quarter of the cases they sampled were cleared by arrest. Researchers who have analyzed cases meeting certain criteria have calculated higher estimates. For instance, of cases that were deemed found, Alderden and Ullman (2012) reported that about 35% of their sample resulted in arrest. Focusing on cases with an identified suspect, some studies have found closer to 50% of cases leading to arrest (Meeker et al., 2021; Spohn & Tellis, 2019; Ylang & Holtfreter, 2020).

Referral for prosecution is another important police decision-making point in case progression. Research generally finds that about 50% or more of sexual assault cases are referred for prosecution (Alderden & Ullman, 2012; Kelley & Campbell, 2013; O’Neal & Spohn, 2017; Wentz, 2020). Focusing on the temporal ordering of arrest and referral, Spohn and Tellis’ (2019) analysis of sexual assault cases from Los Angeles found that about 41% were presented to the prosecution before an arrest was made. This was an important distinction because only about 6% of the cases referred before arrest resulted in charges being filed compared to almost 51% of the cases referred after arrest. Other researchers have reported similar findings in regard to the prevalence of pre-arrest consults in other jurisdictions (e.g., Pattavina et al., 2021).

Legal factors—factors that should have an impact on case progression—are generally found to have the strongest effect. The presence of victim cooperation, physical evidence, victim injury, witnesses, and/or weapon use are legal factors that are frequently found to increase the likelihood of arrest and/or referral (Acquaviva et al., 2022; Alderden & Ullman, 2012; Kelley & Campbell, 2013; Lapsey et al., 2022, 2023; Morabito et al., 2019; Pattavina et al., 2021; Spohn & Tellis, 2019; Wentz, 2020). In fact, Lapsey et al.’s (2022) meta-analysis identified victim cooperation as the most important legal predictor of arrest, followed by physical evidence and the presence of one or more witnesses.

On the other hand, some studies have also identified extralegal factors—factors that should bear no legal influence on a case—that affect police decision-making. One of the most commonly examined is victim credibility, which has been operationalized in a variety of ways to measure victim characteristics and/or behavior. For instance, cases in which the victim suffered from mental health issues (Acquaviva et al., 2022; Morabito et al., 2019) or used drugs or alcohol prior to the assault (Lapsey et al., 2022; O’Neal et al., 2019; Ylang & Holtfreter, 2020) have been found to be less likely to result in arrest. Others have examined victim credibility more broadly or as a scale and found it to significantly affect arrest or referral for prosecution (Alderden & Ullman, 2012; Growette Bostaph et al., 2021; Pattavina et al., 2021). However, there is variation across studies with others finding that victim credibility measures are not significant predictors of outcomes (Lapsey et al., 2023; O’Neal & Spohn, 2017; Wentz, 2020).

The findings on the effects of victim resistance and prompt reporting are also mixed. Alderden and Ullman (2012) and Lapsey et al. (2022) reported that resistance significantly increased the likelihood of arrest, whereas others have not found resistance to predict outcomes (Acquaviva et al., 2022; O’Neal et al., 2019; Spohn & Tellis, 2019). Prompt reporting, which has been conceptualized in varying ways (generally 6–72 hours after the assault), has been found to increase the likelihood of arrest in some studies (Lapsey et al., 2022; Meeker et al., 2021; Wentz, 2020) but not in others (Alderden & Ullman, 2012; Lapsey et al., 2023; Tasca et al., 2012).

Demographics have also been examined as extralegal factors affecting case progression. Victim and/or suspect race/ethnicity are often included in studies of sexual assault case processing and most find that they are not significant predictors (Acquaviva et al., 2022; Spohn & Tellis, 2019; Wentz, 2020). A few exceptions, some of which include unexpected findings, are the odds of arrest increasing for non-White victims (Lapsey et al., 2022; Ylang & Holtfreter, 2020) and decreasing for non-White suspects or certain racial/ethnic victim/suspect dyads (Lapsey et al., 2023; O’Neal et al., 2019). Age is often included as well with some studies finding victim or suspect age to be significant (Lapsey et al., 2022, 2023; Meeker et al., 2021; Pattavina et al., 2021), while other studies have not found age to have an effect on case progression (Alderden & Ullman, 2012; Morabito et al., 2019; Tasca et al., 2012). Victim and suspect sex are rarely examined as many studies limit their samples to cases involving female victims and/or male suspects (Pattavina et al., 2021; Spohn & Tellis, 2019; Wentz, 2020).

Another commonly examined extralegal factor is victim–suspect relationship. Due to the stereotypical notion that a “real rape” occurs between strangers (Burt, 1980; Estrich, 1987), it is often hypothesized that sexual assaults committed by strangers are more likely to progress through the system. Consistent with this hypothesis, Tasca et al. (2012) found that arrest was more likely in stranger cases, and Pattavina et al. (2021) reported that cases involving acquaintances were more likely to receive pre-arrest consults and exceptional clearance. However, Morabito et al. (2019) found that cases involving intimate partners had an increased odds of arrest and Lapsey et al. (2023) reported that arrest was less likely in stranger cases. Still, other studies have not found victim–suspect relationships to impact case processing (Alderden & Ullman, 2012; Meeker et al., 2021; Spohn & Tellis, 2019).

Victim Participation

Victim participation, which can include things such as providing information to investigators and attending interviews, has been found to significantly impact case outcomes (Alderden & Ullman, 2012; Lapsey et al., 2022, 2023; Pattavina et al., 2021; Spohn & Tellis, 2019). Research indicates that about 50%–75% of victims participate in sexual assault investigations (Alderden & Long, 2016; Kaiser et al., 2017; Morabito et al., 2019; O’Neal, 2017). Compared to the body of literature on police decision-making in sexual assault cases, fewer studies have examined predictors of victim participation. Those that have analyzed it as a dependent variable have found that legal factors, such as crime seriousness and evidence increase participation (Anders & Christopher, 2011; Kaiser et al., 2017; Kelley & Campbell, 2013; O’Neal, 2017). However, extralegal factors, such as victim credibility issues, risk-taking, a previous victim–suspect relationship, victim age, and/or victim substance use, have been found to decrease cooperation (Alderden & Long, 2016; Kaiser et al., 2017; Kelley & Campbell, 2013).

Theoretical Background

Two theoretical perspectives have commonly been used to explain sexual assault case outcomes. The focal concerns perspective, originally developed to explain judges’ sentencing decisions (Steffensmeier et al., 1998), has been extended to explain the decision-making of prosecutors and, more recently, police (e.g., Spohn & Tellis, 2019). The main tenets are that decisions are guided by three focal concerns: perpetrator blameworthiness, protection of the community, and practical constraints (Steffensmeier et al., 1998). However, in the absence of information pertaining to these concerns, decision-makers may rely on “perceptual shorthand” (e.g., demographics) (Steffensmeier et al., 1998, pp. 767–768).

Researchers have found that the focal concerns perspective is able to explain some of the variation in sexual assault case outcomes (O’Neal & Spohn, 2017; Spohn & Tellis, 2019; St. George et al., 2022; Wentz, 2020). Most notably, two recent meta-analyses revealed that indicators of practical constraints (e.g., victim cooperation, physical evidence) were the strongest predictors of case outcomes (Lapsey et al., 2022, 2023). Perceptual shorthand in the form of victim–suspect relationship and/or victim and suspect demographics exhibited minimal effects and not always in the expected direction. Measures of victim credibility were not statistically significant predictors in either meta-analysis. The focal concerns perspective has been applied to victim cooperation as well. Kaiser et al. (2017) focused on three focal concerns for sexual assault survivors. They examined the impact of these concerns at various stages and found indicators of all three (e.g., crime seriousness, credibility, risk-taking, strength of evidence) to impact cooperation in differing ways throughout the investigation.

Other studies have focused on feminist theory pertaining to rape myths. Rape myths, which are stereotypical beliefs about sexual violence, minimize the severity of sexual violence, excuse perpetrators, and blame victims (Burt, 1980; Estrich, 1987). These myths support the stereotypical beliefs that false reports of sexual violence are common and that victims are culpable for their victimization, leading to survivors having their credibility doubted and their behavior scrutinized. Essentially, these myths attempt to discount most reports of sexual violence except those that meet the definition of a “real rape” (e.g., stranger perpetrator, injury, credible victim, prompt report) (Burt, 1980; Estrich, 1987). These beliefs also provide an explanation for sexual assault case attrition in that they can lead investigators to doubt the victim’s report (Pattavina et al., 2021; Sleath & Bull, 2017; St. George et al., 2022).

Despite research suggesting that police officers generally have low rape myth acceptance, there is evidence that rape myths permeate sexual assault case processing (Acquaviva et al., 2022; Pattavina et al., 2021; Sleath & Bull, 2017). For example, Venema’s (2016) interviews with police officers revealed that some officers overestimated the occurrence of false reporting and only perceived cases meeting the stereotypical definition of a “real rape” as legitimate. Similarly, St. George and Colleagues (2022) determined that while officers themselves did not necessarily believe rape myths, they knew how rape myths might affect the case during later stages and thus used them as a part of their decision-making. Originally posited to explain prosecutorial decision-making (Frohmann, 1997), the notion of a downstream orientation has been extended to police decision-making as well (Pattavina et al., 2021; Spohn & Tellis, 2019; St. George et al., 2022). That is, police make an educated guess about what they expect the prosecution will do. If they suspect that the case is flawed and will not be successfully prosecuted, they may avoid spending any additional resources on it.

Rape myths can also impact victim participation. Victims may be less likely to participate if they harbor any self-blame, fear they will not be believed, or believe their assault does not meet the definition of a “real rape” (Alderden & Long, 2016; Anders & Christopher, 2011; Patterson et al., 2009). How they are questioned by investigators can also hint at rape myths and discourage cooperation (Alderden & Long, 2016; Anders & Christopher, 2011; Kaiser et al., 2017). For example, if a victim is asked why they did not report sooner or why they were drinking alcohol, it can be interpreted as an expression of doubt or blame, leading the victim to disengage from the process (Alderden & Long, 2016; Morabito et al., 2019).

Focal concerns and feminist theory pertaining to rape myths are complementary in many ways. Feminist theory focuses on the impact of extralegal factors such as victim credibility and victim–suspect relationship to explain case outcomes, but it also incorporates legal factors that function as proxies of “real rape” characteristics (e.g., injury, cooperation). The focal concerns perspective emphasizes legal factors while incorporating perceptual shorthand in the form of demographics, victim–suspect relationship, and victim credibility. However, neither of these perspectives focuses directly on the effect of officer doubt which can impact a victim’s inclination to participate, and subsequently, case attrition.

Officer Doubt

While numerous studies have examined the impact of a variety of measures of legal and extralegal factors, few have directly examined how officer doubt affects case outcomes. Statements or questions regarding a victim’s credibility or certain details of the case may be expressions of officer doubt, but not always. For example, a report may reference that the victim was drinking alcohol, which could be an expression of officer doubt about the victim’s recall or honesty, but it could also be evidence of the victim being unable to consent, thereby strengthening the case. The key is how officers perceive that characteristic.

Thus, this study’s main research question was as follows: How is officer doubt portrayed in sexual assault reports and what is its impact on victim participation and case processing? We controlled for several legal and extralegal factors consistent with the focal concerns perspective and feminist theory pertaining to rape myths to isolate the effects of officer doubt, which we argue is separate from other case characteristics. Thus, the hypothesis for the present study was as follows:

H1: Cases with evidence of officer doubt will have decreased odds of victim participation, arrest, and referral for prosecution while controlling for legal and extralegal factors.

Methods

Data and Sample

Data for this study were gathered from one policing agency in the western U.S. which is considered a mid-sized agency by the International Association of Chiefs of Police (IACP, n.d.). The department has a detective bureau, a victim-witness unit, and a policy requiring the assignment of a trained sexual assault investigator to all sexual assault cases (citation omitted to protect participating agency privacy). The agency serves a city designated as an urbanized area by the U.S. Census Bureau (Ratcliffe, 2022) in which the median income is more than $10,000 below the national average, almost 75% identify as White alone, and just under one-quarter identify as Hispanic/Latino (any race) (U.S. Census Bureau, 2021a, 2021b).

The sampling frame included all sexual offense cases reported to the police department between January 1, 2016 and December 31, 2017. Cases included a range of sexual offenses such as rape, sexual assault with an object, forcible fondling, and sexual battery. In all, 100 cases were selected from 2016 and 100 were selected from 2017 using simple random sampling. The decision to randomly sample 100 cases from each year was made to gather a sufficiently large, probability sample for analysis while also dealing with time and resource constraints. This sample size provided sufficient power to detect medium and large effect sizes (Cohen, 1992).

Although other researchers have focused on cases involving female victims aged 12 years or older (e.g., Spohn & Tellis, 2019) or male suspects (e.g., Wentz, 2020), we chose to include all selected cases in our analysis regardless of victim or suspect demographics. The only exclusion criterion employed was whether police had access to any identifying information about the suspect, such as name, location, or description. Cases in which such information was unavailable were excluded due to this study’s examination of case processing. This resulted in a final sample size of 175 (85 cases from 2016, 90 from 2017). Full access to digital copies of the sampled police reports was granted; they were not redacted in any way.

Measurement

After carefully reading through each sampled report, detailed information was collected and coded for analysis using International Business Machines (IBM) Statistical Package for the Social Sciences Software (SPSS), Version 26.0. One researcher with extensive experience in sexual violence research coded all of the reports. To examine intra-rater reliability, a random sample of 10% of cases (n = 18) was coded for all study variables at two separate times. Kappa coefficients ranged from .609 to 1.0 with the majority (73%) at .723 or higher. Variables with Kappa coefficients below .723 were closely examined for discrepancies and the coding schema was adjusted for consistency using previous research examining sexual assault police reports as a guide.

This study included three dependent variables, each coded dichotomously: victim participation (whether the victim fully participated in the investigation or not by attending interviews, providing information, etc.), arrest (whether an arrest was made), and referral for prosecution (whether police referred the case to prosecutors for review). Unfortunately, we were not able to assess pre-arrest consults as other researchers have (Pattavina et al., 2021; Spohn & Tellis, 2019) because the temporal ordering of these actions was not clear in many reports.

The main independent variable examined was whether officer doubt was expressed in the police report. Officer doubt was conceptualized as any indication that the victim’s statements, credibility, behavior, or other details of the case were met with skepticism by officers and it was coded dichotomously (present = 1, absent = 0). Expressions of officer doubt manifested in several ways: suspicion of false reporting, inconsistencies in statements or behavior, the influence of drugs or alcohol, delayed reporting/help-seeking, lack of injury/corroborating evidence, and lack of victim participation. Some examples of quotes from reports expressing officer doubt are (pseudonyms used in all quotes to protect confidentiality):

I told James [victim] that it looked like he might have been enjoying it.

She [victim] sent him [suspect] 3 back to back messages this morning, clearly soliciting contact from the suspect, and that is not behavior consistent with a rape victim.

Stacy [victim] admitted to having sex for drugs [in the past].

She [victim] did not seek medical attention immediately after the incident, even though she stated there was bleeding that had occurred.

I saw no bruising on Helen’s [victim] face at all that would be consistent with being hit in the face multiple times.

He [victim] did not seem enthusiastic to provide any more information.

A number of indicators of legal and extralegal factors that have been used in previous research to measure focal concerns and rape myths were included as controls. The presence of corroborating evidence (e.g., sexual assault kit, photographs, electronic communications), direct witness(es) to the assault, victim injury, and prompt reporting (i.e., 24 hours or less) were included and coded dichotomously. Weapon use is typically a proxy for the protection of the community or crime seriousness. Unfortunately, we were unable to include it as only five reports mentioned a weapon, resulting in a large standard error. Victim participation, a dependent variable, was also included as a control variable in the arrest and referral for prosecution models. Extralegal factors examined were victim’s sex and age, current/former intimate partner victim–suspect relationship, and whether the victim resisted the attack in any way. We were unable to include suspect sex and age due to multicollinearity with victim sex and age, and victim and suspect race/ethnicity were both plagued by missing data.

Last, a composite scale of victim credibility issues was created as another extralegal factor measure. This scale assessed the presence of factors that have been shown to affect a victim’s credibility. Each item was coded as present in the report (1) or not (0) and summed. The original scale was created using 14 items. However, due to negative or extremely low item-total correlations for victim mental health issues (−.057), reluctance to share details (.003), and previous reports of sexual victimization (.090), the scale was reduced to 11 items. Table 1 displays the frequencies and item-total correlations for the final 11-item scale. The most common credibility issues included the victim’s drug or alcohol use prior to the assault, previous sexual history, having a reason to lie, and trouble remembering details about the assault. Item-total correlations were highest for a history of drug or alcohol abuse, drug or alcohol use prior to the assault, criminal history, and withdrawal of initial consent. While the internal consistency of the scale was somewhat low (α = .553), it is consistent with other research using a summated scale to measure victim credibility in sexual offense cases (St. George & Spohn, 2018). Possible scale scores ranged from 0 to 11 in which higher scores indicated more credibility issues.

Table 1.

Victim Credibility Issues Scale.

Variable	Present in Report n (%)	Item-Total Correlation
Drug or alcohol use prior to assault	63 (36.0%)	.340
Previous sexual history	52 (29.7%)	.257
Has reason to lie	49 (28.0%)	.232
Trouble remembering details	48 (27.4%)	.201
Previous consent with suspect	41 (23.4%)	.194
History of drug or alcohol use	36 (20.6%)	.372
Criminal history	32 (18.3%)	.259
Discrepancies in statements	31 (17.7%)	.165
Other credibility issues	22 (12.6%)	.134
Withdrew initial consent	21 (12.0%)	.259
Homeless/without permanent residence	16 (9.1%)	.129

Results

Descriptive statistics were computed to describe this sample on all variables included in the study. Bivariate analyses were also conducted to examine the relationships among variables. Last, due to the dichotomous nature of the dependent variables, binary logistic regression was used to test this study’s hypothesis.

Table 2 displays descriptive statistics for all variables examined. Officer doubt was present in almost 27% of cases. The victim fully participated in the investigation by attending interviews, providing information, etc. in almost 50% of cases. Only about 22% of cases resulted in arrest compared to 43% of cases being referred to the prosecutor. Corroborating evidence was documented in over half of the cases, one or more direct witnesses to the assault were present in 19%, and the victim was injured in about one-third. Delayed reports were common with only about one-quarter being made within 24 hours of the assault. The majority of victims were female and ranged in age from 2 to 74 with an average of about 21 years (approximately 88% were aged 12 or older and 39% were over 18). Just under one-quarter of cases involved a victim and suspect who were current or former intimate partners and half of the reports noted that the victim verbally or physically resisted the attack. In terms of victim credibility issues, sample scores ranged from 0 to 8 with an average of about two issues noted in each report.

Table 2.

Sample Descriptive Statistics.

Variable/Category	Present in Report n (%)	Mean	SD	Range
Independent variable
Officer doubt	47 (26.9%)
Dependent variables
Victim participation	87 (49.7%)
Arrest	38 (21.7%)
Referred for prosecution	76 (43.4%)
Control variables
Corroborating evidence	103 (58.9%)
Direct witness(es)	33 (18.9%)
Victim injury	56 (32.0%)
Prompt report	48 (27.4%)
Female victim	153 (87.4%)
Victim age		21.03	11.27	2-74
Current/former intimate partner	41 (23.4%)
Resisted	88 (50.3%)
Victim credibility issues		2.35	1.90	0-8

A correlation matrix including all study variables is displayed in Table 3. In regard to the statistically significant correlations with the three dependent variables, evidence collection and victim resistance were positively correlated with victim participation while an intimate partner victim–suspect relationship was negatively related. Victim participation and evidence collection were positively correlated with arrest while officer doubt was negatively related. Referral for prosecution was positively related to victim participation, evidence collection, injury, and resistance, and negatively correlated with victim age. There were some significant correlations among the independent and control variables although all were .35 or less. Notably, officer doubt was significantly correlated with victim credibility issues (.33). However, the relationship was relatively weak, suggesting that while these variables are related, they are separate concepts.

Table 3.

Correlation Matrix.

	1	2	3	4	5	6	7	8	9	10	11	12
1 Particip.	—
2 Arrest	0.31**	—
3 Referral	0.40**	0.57**	—
4 Doubt	0.07	−0.19*	−0.14	—
5 Evidence	0.16*	0.21**	0.24**	0.09	—
6 Witness	0.08	0.07	0.11	0.07	0.17*	—
7 Injury	0.08	0.14	0.19*	0.03	0.35**	0.17	—
8 Prompt	0.00	−0.04	−0.02	0.09	0.28**	−0.00	0.21*	—
9 Female	−0.04	−0.01	−0.02	−0.08	0.10	0.01	0.22*	0.04	—
10 Age	−0.14	−0.12	−0.17*	0.03	0.01	0.00	0.11	0.24**	.03	—
11 Intimate	−0.17*	−0.10	0.01	0.03	0.02	0.01	−0.03	−0.10	0.05	0.08	—
12 Resisted	0.19*	−0.00	0.16*	−0.02	0.14	0.10	0.24**	0.07	0.14	0.18*	−0.07	—
13 Credib.	0.07	−0.06	−0.06	0.33**	0.23**	0.17*	0.20**	0.10	0.16*	0.13	0.16*	0.10

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

The final analyses employed binary logistic regression to test this study’s hypothesis that officer doubt would decrease the odds of victim participation, arrest, and referral for prosecution. In addition to the bivariate correlation matrix discussed above, variance inflation factor (VIF) statistics were calculated to confirm the absence of multicollinearity (VIF ≤ 1.290). The logistic models are displayed in Table 4. Officer doubt was not a significant predictor of victim participation but three control variables were as follows: current or former intimate partners decreased the odds of full participation by about 57% and victim age decreased it by 3%, while resistance increased the odds of participation by 141%. In terms of case outcomes, as hypothesized, officer doubt significantly predicted both arrest and referral for prosecution, decreasing the likelihood of arrest by approximately 81% and referral by 60%. Put another way, cases not containing officer doubt had 415% increased odds of arrest and 146% increased odds of referral. Victim participation and corroborating evidence were also significant with the former showing the largest effect. The likelihood of arrest increased by almost 476% and referral by 565% when the victim fully participated with the investigation. Corroborating evidence increased the likelihood of arrest by about 234% and referral by 179%. Direct witness(es), injury, prompt reporting, victim sex, and victim credibility were not statistically significant predictors in any of the models.

Table 4.

Logistic Regression Models for Victim Participation, Arrest, and Referral.

Variable	ParticipationNaglekerke r² = .152		ArrestNaglekerke r² = .321		ReferralNaglekerke r² = .371
Variable	b	OR	b	OR	b	OR
Officer doubt	0.142	1.153	−1.638	0.194**	−0.901	0.406*
Evidence	0.489	1.631	1.205	3.336*	1.027	2.792*
Witness(es)	0.141	1.152	0.130	1.138	0.344	1.411
Injury	0.028	1.028	0.713	2.039	0.794	2.212
Prompt report	−0.113	0.893	−0.554	0.575	−0.207	0.813
Participation	N/A	N/A	1.750	5.755***	1.894	6.646***
Female victim	−0.609	0.544	−0.260	0.771	−0.170	0.843
Victim age	−0.032	0.968*	−0.018	0.982	−0.037	0.964
Intimate partner	−0.852	0.427*	−0.289	0.749	0.733	2.082
Resisted	0.880	2.410*	−0.693	0.500	0.235	1.265
Credibility issues	0.066	1.068	−0.065	0.937	−0.163	0.849

Note. OR = odds ratio.

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

Discussion

The goal of this study was to determine the impact of officer doubt on victim participation, arrest, and referral for prosecution in sexual assault cases while controlling for relevant legal and extralegal factors implicated by the focal concerns perspective and feminist theory regarding rape myths. We controlled for these factors to isolate the effects of officer doubt while accounting for predictors commonly identified in previous research. The findings have important implications for sexual assault case research, policy, and practice.

This study identified various ways in which officers displayed doubt in written reports (e.g., suspicion of false reporting, inconsistencies in statements or behavior, the influence of drugs or alcohol, delayed reporting, lack of injury, lack of victim participation). Consistent with the hypothesis, officer doubt significantly decreased the odds of arrest and referral for prosecution while controlling for legal and extralegal factors. This suggests that when officers hold doubt about a victim or characteristics of the case, it is a factor in their decision to make an arrest or refer the case for prosecution. It is also possible that officer doubt is representative of a downstream orientation (Frohmann, 1997) in that officers are anticipating how case characteristics will impact prosecution (Pattavina et al., 2021; Spohn & Tellis, 2019).

Contrary to this study’s hypothesis, officer doubt was not a significant predictor of victim participation. It was hypothesized that officer displays of doubt about the victim or case characteristics would lead victims to disengage from the process (Alderden & Long, 2016; Anders & Christopher, 2011; Kaiser et al., 2017; Morabito et al., 2019). However, this study’s focus on written police reports limited the ability to observe how officer doubt played out in interactions. Some of the reports provided direct quotes of conversations while others did not. It is possible that officers harbored doubt but did not directly display it. Consistent with previous research (Alderden & Long, 2016; Anders & Christopher, 2011; Patterson et al., 2009), it is also possible that victims anticipated they would not be believed due to their background, behavior, or other characteristics, so officer doubt had less of an impact on their participation.

The findings of this study suggest that officer doubt is a separate concept from victim credibility factors and other case characteristics. The only control variable it was significantly related to was the victim credibility issues scale and that correlation was relatively weak (.33). Thus, while the results suggest that officer doubt and victim credibility are related, they should be conceptualized and measured as two different concepts. Furthermore, officer doubt was a significant predictor of arrest and referral while the victim credibility scale was not significant in any of the models. We posit that officer doubt is representative of officer cognition, measuring how certain case characteristics are actually interpreted by officers, and subsequently, how that perception impacts case outcomes.

Consistent with previous research (Alderden & Ullman, 2012; Lapsey et al., 2022, 2023; Spohn & Tellis, 2019; Wentz, 2020), the legal factors of victim participation and corroborating evidence were found to significantly increase the odds of arrest and referral, with participation exhibiting the largest effect. Due to practical constraints (focal concerns) or characteristics of a “real rape” (feminist theory), cases with a participating victim and corroborating evidence of the assault were more likely to progress through the system. However, other legal factors representative of “real rape,” suspect blameworthiness, and practical constraints (i.e., victim injury, prompt reporting, and witnesses) that have been found to predict police actions by some researchers (Acquaviva et al., 2022; Lapsey et al., 2022, 2023; Morabito et al., 2019; Pattavina et al., 2021; Spohn & Tellis, 2019) were not significant in this study. In addition, none of the legal factors predicted victim participation which is contrary to other studies (Alderden & Long, 2016; Anders & Christopher, 2011; Kaiser et al., 2017; Kelley & Campbell, 2013; O’Neal, 2017). Given the limited body of research, and disparate findings, further examination of predictors of victim participation in sexual assault investigations is warranted.

The impact of extralegal factors, positioned as perceptual shorthand (focal concerns) or evidence of rape myths (feminist theory) in previous research, garnered some support as well. In fact, two of the three significant predictors of victim participation were that participation was less likely if the suspect was a current/former intimate partner and more likely if the victim resisted. These findings lend support to the notion that victims may be less likely to participate if their assault does not meet the stereotypical definition of a “real rape” (Alderden & Long, 2016; Anders & Christopher, 2011; Estrich, 1987; Patterson et al., 2009). However, other extralegal factors failed to reach significance with the exception of victim age which slightly decreased the odds of referral. It is possible that the presence of these variables is not necessarily the driving factor. Rather, it is how these characteristics are perceived by investigators (i.e., whether they result in officer doubt or not). Future research should continue to investigate the presence of officer doubt, as well as its impact on case progression.

Limitations and Strengths

It is important to consider this study’s limitations. The focus on one jurisdiction in a medium-sized city during a specific timeframe potentially limits the ability to generalize the findings to other time periods or regions of different sizes, cultures, and demographics. The city in which this study was conducted is limited in terms of racial/ethnic diversity (primarily White or Hispanic) and the median income is well below the national average. In addition, the analyzed reports came from a policing agency in which trained investigators are assigned to all sexual assault cases. These characteristics may affect expressions of doubt, how sexual assault cases are handled, and victims’ trust in police and willingness to participate in investigations. Analysis of officer doubt and its impact on sexual assault case processing should be undertaken in areas with more diverse populations and varying police practices pertaining to sexual assault cases.

While most studies of sexual assault case attrition draw their data from written police reports, the method is not without its limitations. Police reports may contain inaccuracies or missing information. This was particularly apparent in the present study in regard to victim and suspect race/ethnicity which were missing from a large portion of reports, making us unable to include these variables in our analyses. Police reports are limited to what officers choose to put in them. Examination of body camera footage or interview recordings would likely provide a more comprehensive depiction of officer doubt in sexual assault cases. In addition, qualitative interviews with victims could illustrate how often they experience officer doubt and how it impacts them. While the information provided in the written report is important, examining interactions between victims, witnesses, and officers would offer rich insights into the presence and impact of officer doubt on victim participation and police actions. In addition, the sample for this study was drawn using random sampling and included all sexual assault cases apart from those in which suspect information was missing. Focusing only on certain types of sexual offenses (e.g., rape) or certain victims (e.g., female victims) as other researchers have done may yield different findings in regard to officer doubt and case processing. Furthermore, the sample size used in this study was not large enough to enable identification of small effect sizes. Larger sample sizes would have more power, potentially resulting in varying findings. Although consistent with other research (St. George & Spohn, 2018), the victim credibility issues scale exhibited somewhat low internal consistency. Including different indicators of victim credibility or examining them individually, rather than as a scale, may change the findings. In addition, the manner in which officer doubt was measured in this study was found to be reliable (Kappa = .769), though other researchers may identify differing ways to measure this important concept.

This study was unable to establish temporal ordering. Many of the reports were unclear about the sequencing of events so further examination of the timing of police actions and victim decisions is warranted. Such analyses may uncover additional information about the manifestation of officer doubt, as well as its impact on victim participation and police decision-making at various stages of the investigation. Last, while the analyses were deemed the most appropriate for the present study, other statistical procedures may result in different conclusions.

Despite its limitations, this study adds to the literature in a number of ways. First, the findings have illustrated the presence and impact of officer doubt in sexual assault case processing, which should continue to be examined in future studies using various methodologies. The analyses suggest that while officer doubt may be influenced by victim credibility and other case characteristics, it is a separate concept representing officer cognition. Second, this study adds to the body of research on predictors of victim participation, arrest, and referral for prosecution. Many of the findings are consistent with previous research but others suggest the need for further investigation into sexual assault case attrition. Third, the present study examined police reports from a medium-sized city in the western U.S. This is important because many other studies on sexual assault case processing focus on larger jurisdictions. Last, the sample was not restricted by victim or suspect sex or age, providing a more comprehensive analysis of the diversity of reported sexual assault cases.

Implications

Officer doubt was present in almost 27% of the cases examined and it significantly impacted police decision-making in terms of arrest and referral for prosecution. This begs the question of exactly how and when officer doubt drives police decision-making, as well as the factors that lead to it. While further research is needed to answer these questions, the frequency with which officer doubt was displayed in the present study is concerning. According to the Department of Justice’s (2015) report on preventing gender bias in sexual assault response, “officers should not make statements or engage in acts that indicate to the victim that they doubt the victim’s credibility” (p. 14) as these actions interfere with the principle of treating all victims with respect and encouraging participation. While police officers are trained to interrogate suspects and gather information for prosecution, training must include skills to interview victims using trauma-informed techniques (Campbell et al., 2020; Franklin et al., 2020; Rich & Seffrin, 2012; Venema, 2016). It should emphasize the realities of sexual violence, the impact of trauma on victim behavior, and clear instruction about when a case warrants arrest and referral for prosecution (Campbell et al., 2020; Franklin et al., 2020; Rich & Seffrin, 2012; St. George et al., 2022; Venema, 2016). Officers should be trained to avoid jumping to conclusions and succumbing to doubt about the victim during the investigation, and most importantly, avoid displaying any doubt directly to the victim or others. While training of this nature does exist and has been found to successfully reduce misconceptions and improve officer understanding of trauma, it may not be available in all locations (Campbell et al., 2020; Franklin et al., 2020; Rich & Seffrin, 2012). Future research should continue to examine the impact of trauma-informed training on officer perceptions and behavior, as well as the feasibility and effectiveness of extending such training to other criminal justice personnel (e.g., prosecutors).

The importance of victim participation in sexual assault cases cannot be understated. It was the strongest predictor of arrest and referral for prosecution in this study and numerous other researchers have identified its importance in case outcomes (e.g., Alderden & Ullman, 2012; Lapsey et al., 2022, 2023; Spohn & Tellis, 2019). Reducing the costs of participating (e.g., the emotional toll of having their credibility questioned) would likely increase victims’ willingness to do so (Kaiser et al., 2017). In addition to police training, research has also shown that having the support of a victim advocate improves the treatment of sexual assault survivors by service providers, as well as decreases stress (Campbell, 2006). Victim advocates can help survivors navigate the system and provide emotional support throughout the process, which has the potential to decrease the negative experiences victims often report and increase the likelihood of participation in the investigation (Campbell, 2006; Patterson & Tringali, 2015).

Efforts to decrease officer doubt and increase victims’ willingness to participate in sexual assault investigations are paramount in improving the treatment of victims, holding perpetrators accountable, and ultimately, reducing sexual assault case attrition. Continued training on the realities of sexual violence and the use of trauma-informed interviewing practices, as well as the increased involvement of victim advocates throughout sexual assault investigations, are two methods that offer promise in the pursuit of these goals.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article.

Funding

The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: This research was funded through a contract with the participating police department which provided the data for this study. Opinions expressed here are those of the authors only.

ORCID iD

Laura L. King

Author Biographies

Laura L. King, PhD, is an associate professor of Criminal Justice at Boise State University where she teaches courses on research methods, statistics, and victimology. Her research focuses on sexual violence, victimization, and gender and crime. Much of her work has involved collaborations with local victim service and criminal justice agencies in Idaho.

Lisa M. Growette Bostaph, PhD, is a professor of Criminal Justice at Boise State University. She earned her degree in Criminal Justice from the University of Cincinnati and her areas of interest are systemic responses to domestic and sexual violence, criminal justice policy, and the use of discretion by policing professionals. Dr. Bostaph has published in numerous peer-reviewed journals.

References

Acquaviva

B. L.

Meeker

K. A.

O’Neal

E. N.

(2022). Blameworthy suspects and “real rape”: Assessing the effects of rape culture-specific suspect culpability factors on the police decision to arrest. Violence Against Women, 28(15–16), 3865–3885. https://doi.org/10.1177/10778012211070314

Alderden

M. A.

Long

(2016). Sexual assault victim participation in police investigations and prosecution. Violence and Victims, 31(5), 819–836.

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

Anders

Christopher

F. S.

(2011). A socioecological model of rape survivors’ decisions to aid in case prosecution. Psychology of Women Quarterly, 35(1), 92–106. https://doi.org/10.1177/0361684310394802

Bureau of Justice Statistics. (2012). Victimizations not reported to the police, 2006–2010. https://bjs.ojp.gov/content/pub/pdf/vnrp0610.pdf

Bureau of Justice Statistics. (2022). Criminal victimization, 2021. https://bjs.ojp.gov/library/publications/criminal-victimization-2021

Burt

M. R.

(1980). Cultural myths and supports for rape. Journal of Personality and Social Psychology, 38(2), 217–230.

Campbell

B. A.

Lapsey

D. S.

Wells

(2020). An evaluation of Kentucky’s sexual assault investigator training: Results from a randomized three-group experiment. Journal of Experimental Criminology, 16, 625–647. https://doi.org/10.1007/s11292-019-09391-0

Campbell

(2006). Rape survivors’ experiences with the legal and medical systems. Violence Against Women, 12(1), 30–45. https://doi.org/10.1177/10778012052777539

10.

Centers for Disease Control. (2022). Sexual violence. Violence Prevention. https://www.cdc.gov/violenceprevention/sexualviolence/index.html

11.

Cohen

(1992). A power primer. Psychological Bulletin, 112(1), 155–159.

12.

Department of Justice. (2015). Identifying and preventing gender bias in law enforcement response to sexual assault and domestic violence. https://www.justice.gov/media/1171861/dl?inline

13.

Estrich

(1987). Real rape. Harvard University Press.

14.

Franklin

C. A.

Garza

A. D.

Goodson

Bouffard

L. A.

(2020). Police perceptions of crime victim behaviors: A trend analysis exploring mandatory training and knowledge of sexual and domestic violence survivors’ trauma responses. Crime & Delinquency, 66(8), 1055–1086. https://doi.org/10.1177/0011128719845148

15.

Frohmann

(1997). Convictability and discordant locales: Reproducing race, class, and gender ideologies in prosecutorial decision-making. Law & Society Review, 31(3), 531–556. https://doi.org/10.2307/3054045

16.

Growette Bostaph

L.M.

King

L. L.

Brady

P. Q

. (2021). How victim credibility and cooperation influence investigative decision-making: Examining DOJ’s gender bias principles for investigating domestic and sexual violence. Policing: An International Journal of Police Strategies and Management, 44(4), 612–627.

17.

International Association of Chiefs of Police (n.d.). Midsize agencies division. https://www.theiacp.org/working-group/division/midsize-agencies-division.

18.

Kaiser

K. A.

O’Neal

E. N.

Spohn

(2017). “Victim refuses to cooperate”: A focal concerns analysis of victim cooperation in sexual assault cases. Victims & Offenders, 12(2), 297–322. https://doi.org/10.1080/15564886.2015.1078864

19.

Kelley

K. D.

Campbell

(2013). Moving on or dropping out: Police processing of adult sexual assault cases. Women & Criminal Justice, 23(1), 1–18. https://doi.org/10.1080/08974454.2013.743365

20.

Lapsey

D. S.

Campbell

B. A.

Plumlee

B. T.

(2022). Focal concerns and police decision making in sexual assault cases: A systematic review and meta-analysis. Trauma, Violence, & Abuse, 23(4), 1220–1234. https://doi.org/10.1177/1524838021991285

21.

Lapsey

D. S.

Campbell

B. A.

Benitez

(2023). Focal concerns theory and case advancement in sexual assault cases: A systematic review and meta-analysis. Journal of Criminal Justice. Advance online. https://doi.org/10.1016/j.jcrimjus.2023.102032

22.

Meeker

K. A.

O’Neal

E. N.

Hayes

(2021). Policing and prosecuting sexual assault: An examination of arrest and initial filing decisions in cases involving adolescent complainants. Justice Quarterly, 38(5), 870–891. https://doi.org/10.1080/07418825.2019.1679863

23.

Morabito

M. S.

Pattavina

Williams

L. M.

(2019). It all just piles up: Challenges to victim credibility accumulate to influence sexual assault case processing. Journal of Interpersonal Violence, 34(15), 3151–3170. https://doi.org/10.1177/0886260516669164

24.

O’Neal

E. N.

(2017). Victim cooperation in intimate partner sexual assault cases: A mixed methods examination. Justice Quarterly, 34(6), 1014–1043. https://doi.org/10.1080/07418825.2016.1244285

25.

O’Neal

E. N.

Beckman

L. O.

Spohn

(2019). The sexual stratification hypothesis: Is the decision to arrest influenced by the victim/suspect racial/ethnic dyad? Journal of Interpersonal Violence, 34(6), 1287–1310. https://doi.org/10.1177/0886260516651093

26.

O’Neal

E. N.

Spohn

(2017). When the perpetrator is a partner: Arrest and charging decisions in intimate partner sexual assault cases-A focal concerns analysis. Violence Against Women, 23(6), 707–729. https://doi.org/10.1177/1077801216650289

27.

Pattavina

Morabito

M. S.

Williams

L. M.

(2021). Pathways to sexual assault case attrition: Culture, context, and case clearance. Victims & Offenders, 16(8), 1061–1076. https://doi.org/10.1080/15564886.2021.1970661

28.

Patterson

Greeson

Campbell

(2009). Understanding rape survivors’ decisions not to seek help from formal social systems. Health & Social Work, 34(2), 127–136.

29.

Patterson

Trengali

(2015). Understanding how advocates can affect sexual assault victim engagement in the criminal justice process. Journal of Interpersonal Violence, 30(2), 1987–1997. https://doi.org/10.1177/0886260514552273

30.

Ratcliffe

(2022). Redefining urban areas following the 2020 census. U.S. Census Bureau.

31.

Rich

Seffrin

(2012). Police interviews of sexual assault reporters: Do attitudes matter? Violence and Victims, 27(2), 263–279. https://doi.org/10.1891/0866-6708.27.2.263

32.

Sleath

Bull

(2017). Police perceptions of rape victims and the impact on case decision making: A systematic review. Aggression and Violent Behavior, 34, 102–112.

33.

Spohn

Tellis

(2019). Sexual assault case outcomes: Disentangling the overlapping decisions of police and prosecutors. Justice Quarterly, 36(3), 383–411. https://doi.org/10.1080/07418825.2018.1429645

34.

St. George

Spohn

. (2018). Liberating discretion: The effect of rape myth factors on prosecutors’ decisions to charge suspects in penetrative and non-penetrative sex offenses. Justice Quarterly, 35(7), 1280–1308. https://doi.org/10.1080/07418825.2018.1529251

35.

St. George

Verhagan

Spohn

. (2022). Detectives’ descriptions of their responses to sexual assault cases and victims: Assessing the overlap between rape myths and focal concerns. Police Quarterly, 25(1), 90–117.

36.

Steffensmeier

Ulmer

Kramer

(1998). The interaction of race, gender, and age in criminal sentencing: The punishment cost of being young, Black, and male. Criminology, 36(4), 763–797.

37.

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(6), 1157–1177.

38.

U.S. Census Bureau (2021a). American Community Survey 1-year estimates data profile, Demographic and housing estimates. Retrieved from https://www.census.gov/programs-surveys/acs/data.html [Table ID not included to protect participating agency privacy].

39.

U.S. Census Bureau (2021b). American Community Survey 5-year estimates subject tables, Selected characteristics of the native and foreign-born populations. Retrieved from https://www.census.gov/programs-surveys/acs/data.html [Table ID not included to protect participating agency privacy].

40.

Venema

R. M.

(2016). Police officer schema of sexual assault reports: Real rape, ambiguous cases, and false reports. Journal of Interpersonal Violence, 31(5), 872–899.

41.

Wentz

E. A.

(2020). Funneled through or filtered out: An examination of police and prosecutorial decision-making in adult sexual assault cases. Violence Against Women, 26(15–16), 1919–1940. https://doi.org/10/1177/1077801219890419

42.

Ylang

Holtfreter

(2020). The decision to arrest in sexual assault case processing: A test of Black’s theory of the behavior of law. Violence Against Women, 26(10), 1141–1163. https://doi.org/10.1177/1077801219862632