Introduction
Women of color face unique harms from gender-based and community violence, but simultaneously face criminalization, enhanced surveillance, and re-victimization by the police when seeking help (Aniefuna et al., Reference Aniefuna, Aniefuna and Williams2020; Bell Reference Bell2016; Brunson and Miller, Reference Brunson and Miller2006a; Hitchens et al., Reference Hitchens, Carr and Clampet-Lundquist2018; Miller Reference Miller2008; Powell and Phelps, Reference Powell and Phelps2021; Richie Reference Richie2012). Research shows that extensive personal and vicarious police contacts engender distrust and cynicism toward the law and law enforcement (Brunson and Miller, Reference Brunson and Miller2006b; Edwards et al., Reference Edwards, Lee and Esposito2019; Rios Reference Rios2011). In turn, these legal orientations are believed to reduce the likelihood that marginalized groups seek help from the police (Desmond et al., Reference Desmond, Papachristos and Kirk2016; Kirk and Papachristos, Reference Kirk and Papachristos2011; Sheppard and Stowell, Reference Sheppard and Stowell2022). Yet evidence suggests that some types of formal help-seeking are particularly high for marginalized women (Anyikwa Reference Anyikwa2015; Bent-Goodley Reference Bent-Goodley2013; Hamberger et al., Reference Hamberger, Ambuel and Guse2007; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019).
In general, research finds that women are more likely than men to seek help for a range of outcomes from reporting crime to the police, to seeking medical attention for physical and mental health issues (Addis and Mahalik, Reference Addis and Mahalik2003; Campagna and Zaykowski, Reference Campagna and Zaykowski2020; Galdas et al., Reference Galdas, Cheater and Marshall2005; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). Although there have been relatively few studies that have examined how police-initiated contacts impact marginalized women’s help-seeking, some evidence suggests that despite holding cynical attitudes towards the police, their immediate needs for safety and services supersede these beliefs (Bell Reference Bell2016; Xie and Lauritsen, Reference Xie and Lauritsen2012; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). This work suggests that marginalized women develop a “situational trust” with police despite underlying reservations (Bell Reference Bell2016).
In contrast, research frequently highlights resistance among marginalized men to seek help from the police and other social institutions (Addis and Mahalik, Reference Addis and Mahalik2003; Galdas et al., Reference Galdas, Cheater and Marshall2005; Lindsey and Marcell, Reference Lindsey and Marcell2012; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). Extensive research has documented that men of color, particularly young Black men, are more likely to be stopped by the police on the street and in traffic stops, experience police misconduct, and be victims of police brutality more than other racial groups (Anderson Reference Anderson1999; Brunson Reference Brunson2007; Brunson and Miller, Reference Brunson and Miller2006b; Carroll and Gonzalez, Reference Carroll and Gonzalez2014; Kramer and Remster, Reference Kramer and Remster2022; Lundman and Kaufman, Reference Lundman and Kaufman2003; Payne et al., Reference Payne, Hitchens and Chambers2017; Rios Reference Rios2011). Men of color frequently express dissatisfaction with the police, with Black men holding the most negative attitudes towards law enforcement (Alberton and Gorey, Reference Alberton and Gorey2018; Brown and Benedict, Reference Brown and Benedict2002; Cheurprakobkit Reference Cheurprakobkit2000; Hurst et al., Reference Hurst, Frank and Browning2000; Peck Reference Peck2015; Taylor et al., Reference Taylor, Turner, Esbensen and Winfree2001; Wu Reference Wu2014). Marginalized men also express deep mistrust of healthcare professionals due to the historical treatment of people of color and fear of being stigmatized (Lindsey and Marcell, Reference Lindsey and Marcell2012). Men more generally are resistant to help-seeking because it opposes traditional gender norms and masculine identities (Addis and Mahalik, Reference Addis and Mahalik2003; Galdas et al., Reference Galdas, Cheater and Marshall2005).
Drawing from prior work, experience with police contacts may impact marginalized men and women differently when it comes to seeking help. To address this hypothesis, we advance the literature by drawing from the Police Public Contact Survey (PPCS) (U.S. Bureau of Justice Statistics 2018, 2020). The PPCS is a national periodic survey of people aged sixteen or older in the United States about their experiences with contacts with the police. Overall, our aim is to understand the extent to which police-initiated contacts can explain variations in help-seeking behavior across demographic groups. Specifically, we examine (1) how does help-seeking vary by race, sex, and class for reporting suspected crimes, calling for non-criminal emergencies, and non-emergency contacts; (2) to what extent does police-initiated contact, operationalized as street stops, traffic stops, and “other” police initiated contacts, modify the relationship between demographic characteristics and help-seeking outcomes; and (3) does the perception of police-initiated contacts as “legitimate” impact help-seeking outcomes? We anticipated that police contacts would be significantly and negatively related to help-seeking, particularly for reporting crimes (Chenane et al., Reference Chenane, Wright and Gibson2020; Desmond et al., Reference Desmond, Papachristos and Kirk2016; Gibson et al., Reference Gibson, Walker, Jennings and Miller2010; Kirk and Papachristos, Reference Kirk and Papachristos2011; Slocum Reference Slocum2018). However, we expected the effect of police contacts to be less impactful for marginalized women’s help-seeking, who, unlike men, face unique forms of gender based and community violence look to law enforcement out of a need for safety amidst the lack of alternative options (Bell Reference Bell2016; Powell and Phelps, Reference Powell and Phelps2021; Xie and Lauritsen, Reference Xie and Lauritsen2012; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019).
Police Contacts as Legal Socialization
Legal socialization is the process in which attitudes and orientations toward the law are developed through personal, vicarious, and cultural influences (Moule et al., Reference Moule, Burruss, Gifford, Parry and Fox2019). Although earlier studies focused on the process of legal socialization toward the goal of obedience to the law, more recent work has examined how structural and cultural contexts shape police behavior, how marginalized groups internalize these interactions, and how perceptions of these experiences impact other outcomes (Bell Reference Bell2016, Reference Bell2017; Moule et al., Reference Moule, Burruss, Gifford, Parry and Fox2019; Powell and Phelps, Reference Powell and Phelps2021). Legitimacy of the law and law enforcement is often grounded in the idea of “procedural justice,” in which the police must treat people fairly during the process of interactions, regardless of the outcomes. When people perceive law enforcement to be legitimate, they are more likely to obey the law, have more positive attitudes toward the police, report crimes and suspicious behavior, and cooperate in other ways (Gau et al., Reference Gau, Corsaro, Stewart and Brunson2012; Sunshine and Tyler, Reference Sunshine and Tyler2003; Tyler and Huo, Reference Tyler and Huo2002; Tyler Reference Tyler2011). However, the procedural justice approach has been critiqued for lacking evidence of causality, failing to adequately address the underlying social conditions that give rise to police contacts and police behavior, and focusing on deference to police authority (Bell Reference Bell2017; Demarest Reference Demarest2021; Nagin and Telep, Reference Nagin and Telep2017). Following a purely procedural justice framework which seeks to increase state power by enhancing obedience does not resolve any of the underlying issues of alienation experienced by marginalized groups (Bell Reference Bell2017).
Police contacts remain central to how marginalized groups view the law. Research shows how repeated police-initiated contacts and limited responses to emergencies erode trust and confidence in the police to provide and/or maintain public safety (Carr et al., Reference Carr, Napolitano and Keating2007; Kirk and Papachristos, Reference Kirk and Papachristos2011). This orientation, known as “legal cynicism,” is the idea that persistent and ubiquitous personal and vicarious experiences with the police shape more deeply held cultural orientations towards the law and its enforcement (Hagan et al., Reference Hagan, McCarthy, Herda and Chandrasekher2018; Kirk and Papchristos, Reference Kirk and Papachristos2011). While legal cynicism isn’t outwardly rejecting the idea of the law and legal system, these perspectives consequently shape the available “toolkit” of understood options, which limit potential actions when safety is in jeopardy (Carr et al., Reference Carr, Napolitano and Keating2007; Kirk and Papchristos, Reference Kirk and Papachristos2011). Legal cynicism is believed to reduce the likelihood that the police are called when there is a crime or other emergency and increase the likelihood of that people will take violence into their own hands to manage disputes (Kirk and Papchristos, Reference Kirk and Papachristos2011).
Legal estrangement builds on the notion of legal cynicism in that attitudes toward the police do not fully capture the structural and procedural conditions that give rise to these perspectives (Bell Reference Bell2017). This perspective addresses procedural injustice, vicarious marginalization, and structural exclusion. Black and economically marginalized communities are not policed in the same way White and more affluent communities are impacted. Marginalized groups are “harshly policed yet underprotected” (Bell Reference Bell2017, Reference Bellp. 2115) and “abandoned” yet “heavily monitored” (p. 2117), also known as the overpolicing-underpolicing paradox. Police policies that focus on minor offenses do more to enhance distrust and leave residents vulnerable to more serious offenses for which response times are slow and residents fear being treated poorly even when they are victims (Bell Reference Bell2017; Nguyen and Roman, Reference Nguyen and Roman2024). Whereas legal cynicism frameworks emphasize how negative experiences with the police make residents resistant to call for help, legal estrangement recognizes that attitudes and experiences with the police do not unilaterally impact the decision-making process for seeking help. Reporting to the police may be shaped by other factors such as the immediate need for help, the perception that an issue is not important enough to the victim to pursue, or complicated feelings about the perpetrator (Bell Reference Bell2017; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). Monica C. Bell (Reference Bell2017) points out how Black and economically disadvantaged people who both chose not to report crimes and those who reported crimes frequently are both decisions shaped by disappointment and frustration with the police. Marginalized victims may have less access to other resources, and although dissatisfied with the police, do not have an alternative solution (Bell Reference Bell2016; Reference Bell2017).
Gendered and Racialized Legal Socialization with Police
Extensive research has found that Black Americans and other racialized groups are more likely to have disproportionate contact with the police (Bell Reference Bell2017; Brunson Reference Brunson2007; Brunson and Miller, 2006; Carroll and Gonzalez, Reference Carroll and Gonzalez2014; Lundman and Kaufman, Reference Lundman and Kaufman2003; Payne et al., Reference Payne, Hitchens and Chambers2017; Rios Reference Rios2011). These experiences are known to cause trauma and lead to harmful health outcomes (Bowleg et al., Reference Bowleg, del Río-González, Mbaba, Boone and Holt2020; Dennison and Finkeldey, Reference Dennison and Finkeldey2021; Jindal et al., Reference Jindal, Mistry, Trent, McRae and Thornton2022; McLeod et al., Reference McLeod, Heller, Manze and Echeverria2020; Smith Lee and Robinson, Reference Lee, Jocelyn and Robinson2019). Black and Hispanic Americans worry five and four times greater than Whites, respectively, that they will be victims of police violence (Graham et al., Reference Graham, Haner, Sloan, Cullen, Kulig and Jonson2020). Much of this work has focused on and prioritized the experiences and attitudes of marginalized men (e.g., Anderson Reference Anderson1999; Bruson Reference Brunson2007; Brunson and Miller, Reference Brunson and Miller2006b; Rios Reference Rios2011). Young men of color are more likely than other groups to be stopped by police, perceive unfair or discriminatory behavior, be the target of police misconduct and violence, and hold less favorable attitudes towards police (Brunson and Miller, Reference Brunson and Miller2006b; Edwards et al., Reference Edwards, Lee and Esposito2019; Rios Reference Rios2011).
Research on the legal socialization of marginalized women and policing has been more limited (Powell and Phelps, Reference Powell and Phelps2021). Poor women of color face interlocking oppressions tied to their intersectional and stigmatized identities (Crenshaw Reference Crenshaw1991; Powell and Phelps, Reference Powell and Phelps2021; Richie Reference Richie2012). As Amber Joy Powell and Michelle S. Phelps (Reference Moule, Burruss, Gifford, Parry and Fox2019) assert, “While Black boys and men do face a higher rate of interpersonal lethal violence, police killings, and imprisonment, this centering of male experiences rendered invisible women of color’s own experiences of racial targeting, harassment, abuse and violence—including their unique gendered vulnerability to private, gender-based, and state violence” (p. 448). Among a smaller but growing body of research, studies show how marginalized Black women face unique challenges and exposure to violence unlike other groups (Aniefuna et al., Reference Aniefuna, Aniefuna and Williams2020; Bell Reference Bell2016; Brunson and Miller, Reference Brunson and Miller2006a; Hitchens et al., Reference Hitchens, Carr and Clampet-Lundquist2018; Miller Reference Miller2008; Powell and Phelps, Reference Powell and Phelps2021; Richie Reference Richie2012).
Like marginalized men, women of color are often criminalized and placed under surveillance (Brunson and Miller, Reference Brunson and Miller2006a; Elliott and Reid, Reference Elliott and Reid2019). However, marginalized women are also uniquely at risk for gender-based violence including sexual assault, intimate partner violence, harassment, and stalking (Miller Reference Miller2008; Powell and Phelps, Reference Powell and Phelps2021; Richie Reference Richie2012). Beth E. Richie (Reference Richie2012) contends that Black women suffer from the “violence matrix” which combines acute harms perpetrated by intimate partners, community members, and the state in physical, sexual, and psychological ways. Powell and Phelps (Reference Powell and Phelps2021) argue that marginalized women of color experience a unique “dual frustration” due to high rates of violence exposure and a desire for protection. In their study women of color were reluctant to seek police assistance because of long and impactful histories of police violence against themselves or others in their community. Experiences with the police revealed a pattern of unpredictability and criminalization when women did reach out for help (Powell and Phelps, Reference Powell and Phelps2021). These systems of power reify inequalities between women of color, White women, and men (Crenshaw Reference Crenshaw1991; Powell and Phelps, Reference Powell and Phelps2021; Richie Reference Richie2012).
Despite concerns over the ability of police to provide protection, fear of criminalization, and the potential of feeling “re-victimized” by police, there is evidence to suggest that marginalized women of color may be limited in their options to respond to the violence matrix and other social problems related to their marginalization (Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). Bell (Reference Bell2016, Reference Bell2017) explains that the legal estrangement of marginalized groups is complex. While poor communities of color both fear and distrust the police, needs arising from marginalization and exclusion limit alternative options (Bell Reference Bell2016, Reference Bell2017). In her study of poor African American mothers, Bell (Reference Bell2016) found that participants had “situational trust” with police in particular situations, despite holding deeply critical attitudes toward them. Zaykowski and colleagues (Reference Zaykowski, Allain and Campagna2019) similarly found that marginalized groups, particularly poor Black women, were more likely than other groups, including affluent White women and poor Black men, to report their own victimization to the police. Women’s decisions to report or not report was rarely due to perceptions of the police and instead focused on the greater, necessitated concern to stop the violence (Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). In addition, Powell and Phelps (Reference Powell and Phelps2021) found that although women of color in heavily policed neighborhoods in North Minneapolis consistently noted their distrust of the police, they also provided instances in which they called the police to respond to neighborhood and interpersonal violence. In these situations, women felt that an immediate response to a threat to their safety was necessary, and the only option.
The impact of police contacts is notably different for White and more affluent women (and men). White Americans, particularly from middle and upper-class backgrounds, are less likely to have police-initiated contact than other groups and hold the most favorable views of the police (Alberton and Gorey, Reference Alberton and Gorey2018; Brown and Benedict, Reference Brown and Benedict2002; Cheurprakobkit Reference Cheurprakobkit2000; Hurst et al., Reference Hurst, Frank and Browning2000; Peck Reference Peck2015; Taylor et al., Reference Taylor, Turner, Esbensen and Winfree2001; Wu Reference Wu2014). Studies show that police are less likely to use force (Kahn et al., Reference Kahn, Phillip Atiba Goff and Motamed2016) and are more respectful (Voigt et al., Reference Voigt, Camp, Prabhakaran, Hamilton, Hetey, Griffiths, Jurgens, Jurafsky and Eberhardt2017) when a suspect is White. Research also shows that police respond to crime scenes faster and exert more effort to follow up with victims of crime when they are White (Howerton Reference Howerton2006). In contrast to Black and Native women, Powell and Phelps (Reference Powell and Phelps2021) found that White women who experienced intimate partner violence maintained positive views of the police and their response to calls for service. Overall, studies show how White women are often portrayed as “ideal victims” and given more social value (Long Reference Long2021; Kleinstuber et al., Reference Kleinstuber, Zaykowski and McDonough2020; Kulig and Cullen, Reference Kulig and Cullen2017; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). While marginalized women may seek help because they lack other alternatives, White women seek help with confidence that they will be treated with respect and their experiences will be taken seriously.
Current Study
The extant research to date illustrates how the process of legal socialization is both racialized and gendered. Studies show how the nature and frequency of police contacts are shaped by multiple marginalities in ways that are both consistent (such as the shared experiences of criminalization among Black men and women) but also unique (such as the “matrix” of vulnerabilities and risk for gender-based victimization experienced by Black women). Prior research highlights the complexity in the relationship between police contact and help-seeking behavior, particularly for marginalized women of color (Bell Reference Bell2016; Powell and Phelps, Reference Powell and Phelps2021). However, to date, research remains limited in this area.
The current study utilizes data from the Police Public Contact Survey [PPCS] to examine (1) How do sex, race, and class shape help-seeking for suspected crimes, non-criminal emergencies, and contacts for non-emergencies; (2) to what extent does police-initiated contact (street stops, traffic stops, and other police-initiated stops) modify the relationship between demographic characteristics and help-seeking outcomes; (3) to what extent are perceptions of police-initiated contacts as legitimate linked to help-seeking? Unlike prior studies, we take advantage of multiple PPCS survey years and prioritize how help-seeking outcomes and the impact of contact vary across the intersectional identities of participants. We anticipated that police-initiated contacts would reduce the likelihood of seeking help for all help-seeking reasons, but that the impact of police contact would be attenuated for women of color who, as research has shown, may prioritize need over existing attitudes and prior experiences (Bell Reference Bell2017).
Methods
Data for this study comes from the Police Public Contact Survey (PPCS), which is a supplement of the National Crime Victimization Survey (NCVS). A subsample of households from the NCVS sample are asked about contacts with police within the previous twelve months approximately every three years. All eligible persons (age sixteen and older) within households are asked to participate in the PPCS. The PPCS begins with screener questions about different types of police-initiated and participant-initiated contacts. Then, participants are asked detailed follow-up questions for contacts involving street stops, traffic stops, and voluntary contacts. Data for the current study combines the 2018 and 2020 collection years. Response rates for the PPCS ranged from 74.7% to 75.7% from the NCVS original samples. Including the initial sampling of the NCVS master sample, the overall PPCS response rates ranged from 50.3% to 53.6%. The combined dataset included 199,615 cases of participants who had completed PPCS interviews.
Dependent Variables
There were three dependent variables for help-seeking: Reporting a Crime, Called for an Emergency, and Non-Emergency Contact. Participants were asked a series of screening questions for each variable: (1) “Have you reported any kind of crime, disturbance, or suspicious activity to the police in the past twelve months?” (2) “Have you reported an EMERGENCY that was NOT a crime to the police? These could include medical emergencies, or a traffic accident that you witnessed but were not involved in.” (3) “In the past twelve months, have you contacted or approached police for NON-EMERGENCY assistance such as asking directions, custody enforcement, court orders, or any other non-emergency situation?” Participants who responded “yes” were coded as 1, while “no” was coded 0.
Independent Variables
Participants’ demographic characteristics were from the National Crime Victimization Survey are included in the publicly available PPCS dataset as recoded variables. Race/Ethnicity was measured as Non-Hispanic White Only [reference]; Non-Hispanic Black Only; Hispanic; Non-Hispanic Asian Only; and Non-Hispanic Other Race. Participants who were in the “Other Race” group include Asian/Asian Americans, Native Hawaiian and Pacific Islanders, American Indians/Alaska Natives and participants who were defined as multiracial. Sex was measured as Female/Male. Age categories included 16-24, 25-44, 45-64, and sixty-five or older. To make the 2018 and 2020 datasets comparable it was not possible to evaluate households with greater economic disadvantage ($24,999 and under) as this information was only contained in the 2018 dataset. Therefore, Income was measured as household incomes that were $49,999 or less and $50,000 or more.
Police-Initiated Contact was measured by whether someone had one of three types of contacts: traffic stop, street stop, or other police-initiated contact. For Traffic Stops, participants were asked “Have you been pulled over by the police while driving a motor vehicle, NOT including any driving violations captured by camera and ticketed by mail?” For Street Stops, participants were asked “In the past twelve months, have you been stopped by the police while standing, walking, or sitting in a public place or sitting in a parked vehicle? This could include being stopped because the police were looking for information, were asking about a crime or disturbance, suspected you of something, or if they stopped you for some other reason.” Other police contacts were measured by the question: “Other than what you told me about, have you contacted the police or did they initiate contact with you for any other reason?” Police-initiated contacts from responding to traffic accidents were asked separately and excluded from this variable. If participants had any of these three contact types, they were coded 1 = police-initiated contact, 0 = no police-initiated contact. Only “other” police initiated contacts had missing information, and this was for three participants. Since they had no other contacts with police, we coded these cases as “0,” indicating no police initiated contacts.
Participants who experienced a traffic stop or street stop were asked a follow-up question, “Would you say the police had a legitimate reason for stopping you?” For those people who had more than one traffic or street stop, follow-up questions were asked regarding the most recent stop. Legitimacy was then measured in two variables (one for each type of stop) as participants who had no contact, had contact and perceived it to be legitimate, had contact but perceived the stop not to be legitimate, and had contact but assessment of legitimacy was unknown. Assessments might be unknown if the stop occurred, but was not the only or most recent stop, or if the participant “didn’t know,” or data were missing. Missing data was only problematic for perceptions of legitimacy, and therefore to retain cases we included the additional category for “legitimacy unknown.”
Analytic Approach
Each survey year was analyzed for compatibility across questions. The PPCS has undergone multiple revisions in both survey administration, question wording, and question ordering which limits looking at longer term trends and combining multiple years of data (Berzofsky et al., Reference Berzofsky, Ewing, DeMichele, Langton, Hyland and Davis2017; Ewing and Berzofsky, Reference Ewing and Berzofsky2015). Next, descriptive statistics of the sample were estimated using person weights and survey estimation methods to account for complex structure and sampling method of the dataset (Shook-Sa et al., Reference Shook-Sa, Couzens and Berzofsky2015). Multivariate analyses used a series of logistic regressions to predict the three dichotomous help-seeking outcomes.
First, models were estimated with just participant characteristics to establish a baseline of help-seeking across social groups. Because police experiences and help-seeking behavior may vary intersectionally (e.g., Bell Reference Bell2016; Powell and Phelps, Reference Powell and Phelps2021; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019), each model examined whether demographic interactions were in fact significant. Next, police-initiated contacts were added to the models to observe their impact on help-seeking and whether they modified the association between demographic characteristics and help-seeking. Sensitivity analyses were conducted for different measures of police-initiated contacts including the number of contacts and measuring each type of contact separately. These results (available by request) showed that findings were substantively the same across measurement types. The final models examined whether perceptions of legitimacy influenced help-seeking outcomes.
Results
Sample Characteristics
Table 1 presents the unweighted sample characteristics. Few participants had police contact through participant-initiated or police-initiated stops. Among participants in the sample, 6.6% had reported a crime or disturbance, 3.5% reported a non-crime emergency and 3.3% contacted the police for a non-emergency. About one in ten participants experienced a police-initiated contact (8.9%). About 30% of those stopped (for both street and traffic stops) perceived it to be for illegitimate reasons. The majority of the sample was White (non-Hispanic) making up 62.8% of the sample followed by Hispanic (16.8%), non-Hispanic Black (12.0%) and non-Hispanic Other and Multiracial groups (8.4%). Fifty-three percent of the sample was female and had household incomes above $50,000 (58.4%). Age distributions in the sample were 14.5% (16-24 years old), 33.3% (25-44 years old), 31.9% (45-64 years old) and 20.3% for age sixty-five and older. About half of the sample came from the 2020 PPCS year (50.2%). All bivariate relationships between race and the other variables are significant [p<.001], with the exception of survey year, which is not significant.
Sample Characteristics
Help-seeking by Reason for Contact
Table 2 presents the likelihood of seeking help by reason for contact. Across all models, non-White racial and ethnic groups were less likely to seek help relative to White participants. In addition, females were more likely to seek help than males. Help-seeking was also higher for participants in the younger age groups relative to those sixty-five and older and less likely for respondents in the 2020 survey year relative to 2018. Across each of the models, as shown in the interaction terms, Hispanic participants from higher income households and women from lower income households had an elevated probability of seeking help.
Help-seeking by Reason for Contact
N=199,615 unweighted cases. References: White (non-Hispanic), Age sixty-five and older. Black and Other Race groups are non-Hispanic.
*** p<.001.
** p<.01.
* p<.05.
There were also some unique findings within each model. For contact because of a suspected crime or other disturbance, Black and Hispanic women had an elevated probability of seeking help (Exp[B] = 1.21 and Exp[B] = 1.19, p<.05) relative to their male counterparts. Income did not have a main effect on seeking help for crimes. However, higher income was associated with a greater likelihood of seeking help for non-crime emergencies and non-emergencies (Exp[B] = 1.18 and Exp[B] = 1.19, p<.05).
Influence of Police Initiated Contact on Help-Seeking
Table 3 shows the results of the association between police-initiated contacts on help-seeking. Overall, police-initiated contact was associated with higher likelihoods of calling the police for a crime, emergency, and non-emergency. In addition, police-initiated contact did not modify the main effects of race, sex, or income. Interaction terms with race, sex, and police contact show that Black participants (Exp[B] = 1.59) had an elevated likelihood of seeking help for crimes, but lower when the Black participant was female (Exp[B] = 0.59). Hispanic participants who had police-initiated contact were more likely to seek help that White participants for non-crime emergencies and non-emergencies (Exp[B] = 1.50 and Exp[B] = 1.73).
Influence of Police Initiated Contact on Help-seeking
N=199,615 unweighted cases. References: White (non-Hispanic), Age sixty-five and older. Black and Other Race groups are non-Hispanic. PIC=Police Initiated Contact (traffic stops, street stops, and other police initiated stops).
*** p<.001.
** p<.01.
* p<.05.
Influence of Perception of Perceptions of Legitimacy
Finally, Table 4 presents the results of the relationship between perceptions of legitimacy and help-seeking. Consistent across all models, relative to participants who had a legitimate police-initiated traffic stop, the odds of seeking help were no different for participants who had illegitimate traffic stops. In addition, participants who had not experienced a street stop were less likely to report all outcomes relative to those who were stopped but perceived these experiences as legitimate. Unlike traffic stops, participants who experienced a street stop and perceived it to be illegitimate were less likely to seek help. These differences were significant for reporting crime or disturbances to the police (Exp[B] = 0.50, p<.05). For reporting crimes, there was no difference between people who were not stopped and perceived the stop to be legitimate. Unknown perceptions of legitimacy were positively related to reporting in each model. Accounting for perceptions of legitimacy did not substantively moderate the relationship between the demographic variables and the outcomes (relative to results in Table 2 and Table 3).
Influence of Perceptions of Legitimacy on Help-seeking
N=199,615 unweighted cases. References: White (non-Hispanic), Age sixty-five and older, Legitimate Stop (stopped, but perceived contact to be legitimate). Black and Other Race groups are non-Hispanic.
*** p<.001.
** p<.01.
* p<.05.
Discussion
Research on legal socialization suggests that police-initiated contacts, particularly contacts perceived as illegitimate, would increase feelings of legal cynicism—the belief that the police are untrustworthy and unwilling to help (Bell Reference Bell2017; Carr et al., Reference Carr, Napolitano and Keating2007; Kirk and Papachristos, Reference Kirk and Papachristos2011). Consequently, people who hold cynical beliefs would be less likely to seek help from police (Kirk and Papachristos, Reference Kirk and Papachristos2011). However, recent theoretical and empirical works argue that marginalized groups, most notably Black Americans who have high levels of legal cynicism, may in fact be as likely or more likely to seek help than other groups (Bell Reference Bell2016, Reference Bell2017; Hagan et al., Reference Hagan, McCarthy, Herda and Chandrasekher2018; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019).
In an effort to assess these disparate findings, the current study used data from the PPCS (2018 and 2020) to first examine how sex, race, and class shape help-seeking for crimes, non-criminal emergencies, and non-emergencies. Next, we analyzed whether police-initiated contact and perceptions of legitimacy were related to help-seeking outcomes and if these contacts modified the relationship between demographic characteristics and these outcomes. Overall, we found some support that people of color (Black, Hispanic, and “Other” Race) were less likely to seek help from police for all types of help (crime, emergency, and non-emergency). These findings were consistent with prior work which argues that people of color are hesitant to seek help from police (e.g., Brunson and Miller, Reference Brunson and Miller2006b). In addition, we see consistency with the increased likelihood of women to seek help (Addis and Mahalik, Reference Addis and Mahalik2003; Campagna and Zaykowski, Reference Campagna and Zaykowski2020; Galdas et al., Reference Galdas, Cheater and Marshall2005; Powell and Phelps, Reference Powell and Phelps2021; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019). However, we also found some support that some Black and Hispanic women had slightly elevated odds of seeking help for reporting suspected crimes or disturbances, but not other types of help-seeking. These findings parallel the work by Bell (Reference Bell2016) who found that despite holding cynical views of the police, there were situations in which marginalized women of color sought help. In addition, people with higher income, regardless of race and sex, were more likely to seek help for non-crime emergencies and non-emergencies but were no different in reporting a crime or disturbance, suggesting that affluent people may be more comfortable seeking help from the police including for non-urgent matters.
Where we saw departures from our expectations, we found that police-initiated contacts were associated with increased odds of help-seeking. We also observed that Black and Hispanic participants had elevated help-seeking when they had a police-initiated contact. We can look to Bell (Reference Bell2016, Reference Bell2017) in determining how to begin to explain such contradictions. She argues that Black communities suffering economic marginalization do not reject but rather believe in and support the law in a general sense, but such obedience can exist in tandem with legal cynicism, particularly if we understand that necessity does not equate to or imply legitimacy. Marginalized peoples could simply be likelier to have both police and person-initiated contacts because of broader structural forces (Bell Reference Bell2017). Elevated odds of help-seeking for Black and Hispanic women provides additional evidence of women adopting “situational trust” when necessary (Bell Reference Bell2016). It also may reflect the complexity of the “violence matrix” in which women may face harms from the state by reporting, and threats from intimate partners or members of the community (Richie Reference Richie2012).
We found partial support for our predictions about the relationship between perceptions of legitimacy and help-seeking. Participants who experienced a street stop and perceived the stop to be illegitimate were less likely to report a crime or disturbance than participants who had legitimate stops. Our finding that illegitimate street stops hinder help-seeking suggests that perceptions of particular police-initiated contacts do matter (Gau et al., Reference Gau, Corsaro, Stewart and Brunson2012; Sunshine and Tyler, Reference Sunshine and Tyler2003; Tyler and Huo, Reference Tyler and Huo2002; Tyler Reference Tyler2011). However, perceptions of the legitimacy of traffic stops were not related to help-seeking for crimes, emergencies, and non-emergencies. The insignificant impact for traffic stops on help-seeking may be due to the types of stops that people of color experience. Charles R. Epp and colleagues (Reference Epp, Maynard-Moody and Haider-Markel2014) found that “investigatory” traffic stops were far more harmful in eroding trust and willingness to call the police for help than “traffic safety” stops. They found that investigatory stops which were more common among African Americans than White drivers, were perceived as intrusive and subjective, and were also more likely to lead to punitive outcomes (Epp et. al., Reference Epp, Maynard-Moody and Haider-Markel2014).
Although our work notably suggests revisiting the relationship between legal cynicism and help-seeking, we also recognize there are notable limitations. First, we are unable to capture the frequency and nature of situations that might warrant seeking help. In other work (e.g., Zaykowski et al., Reference Zaykowski, Allain and Campagna2019), outcomes are assessed by how many victimizations one experiences, and then of those, what proportion were reported. We do not have any variables that are able to document how often participants perceived a crime might be occurring or how many emergencies occurred. We are also not able to confirm the causal order of police-initiated contacts and help-seeking. It is possible that in some cases, police contact occurred after help was sought. In addition, how we measure legitimacy in this study was fairly limited. It is possible that a multifaceted measure that captures the more complex aspects of the concept may provide additional insights to our findings. Finally, legal cynicism addresses a cultural frame across broader contexts (e.g., across neighborhoods). While the PPCS has some information on urbanicity, these variables are not in the publicly available files for 2018 and 2020, which were used in this study. Furthermore, the impact of police contacts and perceptions of legitimacy may be developed over time and within social contexts rather than within isolated incidents.
In summary, although past research has often suggested that police contacts reduce the odds of help-seeking as they are traditionally argued to diminish perceived legitimacy, we find that this relationship might be more complex. While participants who experienced an illegitimate street stop were less likely to seek help from the police, the addition of perceived legitimacy to our models did not modify the relationship between the demographic characteristics and help-seeking. This suggests that hesitancy to contact the police must be explained by more than one negative experience. In other words, it is likely that feelings of legal estrangement and cynicism towards the law is embedded much deeper within communities of color (Bell Reference Bell2017). White women were more likely than other groups to contact police, suggesting that they are not only given more value in society (Long Reference Long2021; Kleinstuber et al., Reference Kleinstuber, Zaykowski and McDonough2020; Kulig and Cullen, Reference Kulig and Cullen2017; Zaykowski et al., Reference Zaykowski, Allain and Campagna2019) but may also be more likely to take advantage of this privilege (e.g., Powell and Phelps, Reference Powell and Phelps2021). Further research is needed to explore the relationships between police contact and marginalized groups, and how help-seeking should be understood through an intersectional lens which recognizes the different and simultaneous forms of marginality.
Open access
