Abstract
Despite compromising women’s health and safety, intimate partner violence (IPV) is among the most underreported crimes, and our understanding of factors that drive police reporting by race/ethnicity is underdeveloped. The purpose of this study is to examine racial/ethnic differences in self-reporting IPV to police.
Race/ethnicity-stratified models identified predictors of reporting IPV to police among recent, female survivors (n=898) in the National Crime Victimization Survey (NCVS; 2011–15). Focus groups (n=3) with recent survivors (n=19) in Baltimore, MD (2018) contextualized results.
Black women in the NCVS were twice as likely to report IPV to police relative to White women (AOR=2.02, 95% CI: 1.01–4.15). In race/ethnicity-stratified models, police reporting significantly increased with increasing age between 18 and <35 years (AOR=1.18, 95% CI: 1.05–1.33) for Black women, and with IPV-related injury for Black (AOR=2.51, 95% CI: 1.10–5.71) and Hispanic women (AOR=2.87, 95% CI: 1.22–6.71); Hispanics with less than a high school education were least likely to report (AOR=0.24, 95% CI: 0.07–0.91). Focus groups explained racial/ethnic influences on reporting including a culture of silence and discrimination, socioeconomic status, and social desirability.
We identified influences on reporting IPV to police that vary by race/ethnicity using national data in context to an urban environment. Results demonstrate the need to enhance equity in survivor’s health and public safety through training and organizational change.
Keywords: Intimate partner violence, police, race, community health
INTRODUCTION
More than one-third of American women will experience violence from an intimate partner [1]. Intimate partner violence (IPV) is associated with poor health, homelessness, and homicide among women [1–4], contributing to over 50% of homicides among women and racial/ethnic disparity [2, 5]. Despite compromising women’s health and safety, IPV is among the most underreported crimes.
Research to date shows that IPV incident characteristics (e.g., injury, relationship with perpetrator) and a constellation of reporting considerations (e.g., familial bonds, financial stability) can facilitate or hinder reporting following IPV [6–10], as experiences of IPV are often cyclic with a graduated increase in severity and injury risk [11]. Reporting IPV to police can yield safety and health benefits such as guided safety planning and linkage to medical treatment and IPV resources victims, restricted firearm access for perpetrators, violence interruption, and perceived safety among victims [12–15]. However, in some instances, engaging with police for IPV may not ensure safety as leaving an intimate partnership can be the most dangerous point in the abusive relationship [11], and reporting to police may result in unintended consequences such as arrest of the victim [16] or economic insecurity resulting from perpetrator arrest. Despite the potential for harm, understanding patterns of police reporting following IPV remains an important priority at the interface of public policy and public health.
Our understanding of factors that drive police reporting by race/ethnicity is underdeveloped [17, 18]. Factors like offender reprisal, past experiences with police, and community-reinforced perceptions of law enforcement may influence women’s police engagement for IPV [18, 9, 10]. Reasons for non-report of abuse, specifically among Black women, include institutional racism, self-blame, and socio-cultural norms that mandate protecting same-race perpetrators from retribution (e.g., incarceration), non-disclosure of private matters, and stereotypical strength [19, 20]. Social norms that minimize abuse, especially abuse perpetrated by a known individual, and abuse-related stigma can discourage reporting [21]. Women who experience IPV or sexual violence are two to three times more likely to experience abuse or a neglectful response from law enforcement; such experiences are more prevalent among Black and Latina IPV/sexual violence victims [22]. The intersectionality of race/ethnicity and gender-specific tensions compound police reporting for IPV [18, 23].
Racial/ethnic disparities in police engagement and response are increasingly presented in national media and research, underscoring historical mistrust in the justice system and compromised the health of minorities [18, 24–28]. Discriminatory police practices have spill-over effects [26, 29]. In recent national surveillance, one in three Black respondents forwent seeking police help to avoid unjust treatment—approximately 50% reported personally experiencing discrimination from police in their lifetime [18]. Furthermore, poor mental health days increased significantly among Black Americans following the killing of an unarmed Black American by police in their respective states; no significant effect was observed among White Americans [27].
It may be counterintuitive that national surveillance, including the National Crime Victimization Survey (NCVS) [20], show that Black women are more likely to contact police for IPV [9, 10]. Despite compromised relationships, particularly with marginalized and minority communities, police officers remain front-line to our response to violence against women given their proximity to IPV victims [30]. Thus, it is essential to gain a better understanding of race/ethnic-based drivers of police reporting in context.
Past analyses identified racial/ethnic differences in police reporting but missed the opportunity to understand why these differences exist. We extend prior NCVS analyses [17] to include race-stratified analyses as well as triangulation of quantitative and qualitative findings to clarify socio-structural influences on police reporting.
METHODOLOGY
This sequential mixed-methods study examines racial/ethnic differences in self-reporting IPV to the police (“reporting”) using the Bureau of Justice Statistics’ NCVS and contextualizes these findings via reflective, participatory focus groups with IPV survivors in Baltimore, MD. The City of Baltimore is a place where a compromised relationship between law enforcement and the greater community has been well-documented—this setting bolsters our study design [31, 32, 27].
Quantitative Methodology: NCVS
The NCVS is a nationally-representative, cluster-randomized sample of US households [33] designed to capture self-reported crimes within a six-month recall period. We pooled data (January 2011-December 2015) to increase sample size. Analyses include female respondents (n=898, 53.8% of the full sample) with assaults perpetrated by a current/former intimate partner. In cases of multiple offenders, incidents were categorized as IPV if any of the offenders were an intimate partner.
Self-reporting to the police is the primary outcome [34]. Additional covariates include respondent age, education, household income, residence (urban, rural, public housing), abuse (sexual vs. non-sexual IPV), relationship with offender, incident location, offender substance use, weapon presence, injury experiences and severity, and respondent’s self-defense [2, 11, 35].
Race/ethnicity-stratified descriptive statistics for demographic and incident characteristics were generated. We calculated weighted sample proportions with series-adjusted victimization weights [36]. The weighted proportion of women who reported IPV to police was calculated separately at each level of each covariate and stratified again by race/ethnicity. We assessed differences using chi-square or fisher’s exact.
Logistic regression assessed the odds of self-report by race/ethnicity and other covariates. First, covariates were assessed independently in unadjusted bivariate models. Age was modeled as a linear spline with one knot at age 35; other variables are categorical. Next, stepwise deletion generated adjusted models for the full sample, and within each race/ethnicity strata. Initially, models included all covariates; covariates with the highest p-value were dropped individually until all covariates had a p-value <0.1. Based on focus group discussions, we included income in our models. Household income had a high degree of missingness (~25%)— “missing” was imputed to minimize dropped cases for a complete case analysis. Variance was estimated using Taylor Series linearization sampling with replacement to account for the complex sampling design [36]. Analyses were conducted using SAS 9.4 [37] and SAS-Callable SUDAAN 11 [36].
Qualitative Methodology: Focus groups with IPV survivors to interpret key NCVS results
We conducted focus groups (n=3) in Baltimore, MD (May 2018). Eligible participants were 18 years or older and experienced past-year IPV. We recruited women from support programs and Facebook with flyers. Participants (n=19) received $25 and an IPV/SV support services list.
The focus groups lasted for 60–90 minutes. Consistent with explanatory, reflective participatory methodology [38], we presented NCVS results in lay terms using [38] infographics (Figure 1). Participant interpretation is a fundamental participatory research tool. A semi-structured guide facilitated participant interpretation. We emphasized potential explanatory factors and the potential role of social and structural factors on reporting. Discussions were audio-recorded and transcribed verbatim. We read transcripts for thematic analysis and iteratively refined the preliminary codebook. Peer debriefing throughout data collection, an audit trail during analysis, and dual coding with discussion of discrepancies enhanced reliability [39]. We triangulated qualitative and quantitative results to assess the complementarity of data sources by evaluating full/partial agreement, disagreement, silence, of themes [40].
Figure 1.
An infographic presented to IPV survivors during focus group sessions.
RESULTS
Female IPV victims from the NCVS were primarily 18–44 years (mean=34 years), from urban environments (84%), with at least some post-secondary education (59%). Nearly two-thirds were White; remaining women were Black (15%), Hispanic (11%), or other (11%) (Table 1).
Table 1.
Demographic and incident characteristics (unweighted N and weighted %) of female IPV survivors by race/ethnicity*, in the National Crime Victimization Survey
| Total, N=898 | White, N=560 (63%) | Black, N=148 (15%) | Hispanic, N=126 (11%) | Other, N=64 (11%) | P-value** | |
|---|---|---|---|---|---|---|
| Age | ||||||
| <18 years | 24 (3%) | 15 (4%) | 3 (2%) | 1 (1%) | 5 (2%) | <0.001 |
| 18–24 years | 175 (21%) | 112 (22%) | 23 (17%) | 30 (28%) | 10 (15%) | |
| 25–34 years | 308 (32%) | 178 (27%) | 42 (26%) | 57 (44%) | 31 (62%) | |
| 35–44 years | 201 (24%) | 123 (25%) | 47 (36%) | 26 (19%) | 5 (9%) | |
| 45–54 years | 137 (17%) | 94 (20%) | 24 (18%) | 9 (7%) | 10 (13%) | |
| 55–64 years | 16 (1%) | 14 (2%) | 0 (0%) | 1 (1%) | 1 (<1%) | |
| Education | ||||||
| Did not graduate high school | 147 (20%) | 81 (18%) | 23 (22%) | 31 (33%) | 12 (13%) | 0.002 |
| High school diploma | 250 (22%) | 155 (22%) | 57 (31%) | 31 (23%) | 7 (5%) | |
| Any post-secondary education | 495 (59%) | 320 (61%) | 68 (47%) | 62 (44%) | 45 (81%) | |
| Income | ||||||
| < $10,000 | 151 (22%) | 96 (19%) | 28 (37%) | 20 (22%) | 7 (17%) | 0.009 |
| $10,000 - $19,999 | 136 (19%) | 75 (18%) | 27 (20%) | 25 (29%) | 9 (13%) | |
| $20,000 - $34,999 | 155 (23%) | 91 (21%) | 34 (26%) | 23 (20%) | 7 (29%) | |
| $35,000 - $74,999 | 151 (22%) | 102 (25%) | 18 (16%) | 19 (20%) | 12 (15%) | |
| $75,000+ | 92 (14%) | 70 (17%) | 1 (1%) | 8 (8%) | 13 (26%) | |
| < 2015 poverty line | 283 (41%) | 168 (36%) | 60 (62%) | 42 (53%) | 13 (26%) | 0.006 |
| Living environment | ||||||
| Urban | 748 (84%) | 433 (79%) | 142 (97%) | 124 (98%) | 49 (81%) | <0.001 |
| Rural | 150 (16%) | 127 (21%) | 6 (3%) | 2 (2%) | 15 (19%) | |
| Public housing | 49 (8%) | 24 (8%) | 17 (12%) | 8 (7%) | 0 (0%) | 0.012 |
| Abuse type | ||||||
| Non-sexual | 807 (88%) | 514 (90%) | 140 (96%) | 103 (80%) | 50 (78%) | 0.028 |
| Sexual | 91 (12%) | 46 (10%) | 8 (4%) | 23 (20%) | 14 (22%) | |
| Relationship with Offender | ||||||
| Spouse | 289 (35%) | 183 (40%) | 41 (20%) | 53 (40%) | 12 (23%) | 0.001 |
| Ex-spouse | 83 (9%) | 64 (10%) | 6 (12%) | 11 (7%) | 2 (2%) | |
| Boy/girlfriend or ex-boy/girlfriend | 526 (56%) | 313 (50%) | 101 (68%) | 62 (53%) | 50 (75%) | |
| Incident location | ||||||
| Respondent’s home | 597 (65%) | 363 (64%) | 112 (75%) | 77 (61%) | 45 (62%) | 0.286 |
| Near own home | 101 (14%) | 66 (14%) | 13 (11%) | 17 (14%) | 5 (18%) | |
| At or near friend/neighbor/relative’s home | 101 (9%) | 70 (10%) | 10 (7%) | 17 (14%) | 4 (3%) | |
| Other | 99 (12%) | 61 (12%) | 13 (7%) | 15 (11%) | 10 (16%) | |
| Offender substance use (yes) | 386 (58%) | 243 (56%) | 62 (58%) | 43 (47%) | 38 (75%) | 0.249 |
| Weapon present (yes) | 121 (14%) | 66 (11%) | 32 (23%) | 12 (11%) | 11 (22%) | 0.062 |
| Firearm | 38 (4%) | 25 (4%) | 8 (7%) | 3 (2%) | 2 (1%) | 0.034 |
| Sharp/edged weapon | 41 (4%) | 19 (2%) | 14 (10%) | 5 (4%) | 3 (3%) | 0.051 |
| Blunt weapon | 22 (4%) | 11 (4%) | 7 (3%) | 2 (1%) | 2 (7%) | 0.488 |
| Other weapon or type unknown | 15 (2%) | 9 (1%) | 3 (3%) | 1 (1%) | 2 (7%) | 0.562 |
| Self-defense | ||||||
| -used physical force against offender | 64 (9%) | 36 (6%) | 9 (5%) | 8 (5%) | 11 (33%) | 0.381 |
| -resisted or captured offender | 183 (20%) | 116 (19%) | 30 (18%) | 22 (18%) | 15 (32%) | 0.684 |
| -scared or warned off offender | 153 (18%) | 96 (16%) | 26 (21%) | 18 (15%) | 13 (26%) | 0.696 |
| -persuaded or appeased offender | 138 (15%) | 94 (15%) | 19 (11%) | 18 (13%) | 7 (15%) | 0.773 |
| -escaped or got away | 133 (18%) | 75 (17%) | 24 (16%) | 24 (18%) | 10 (25%) | 0.878 |
| -got help or gave alarm | 145 (15%) | 86 (15%) | 30 (18%) | 23 (15%) | 6 (10%) | 0.657 |
| -reacted to pain or emotion | 78 (7%) | 48 (7%) | 14 (8%) | 7 (8%) | 9 (8%) | 0.974 |
| -other | 37 (5%) | 24 (6%) | 12 (7%) | 0 (0%) | 1 (1%) | 0.004 |
| -any | 515 (60%) | 324 (60%) | 88 (60%) | 71 (52%) | 32 (66%) | 0.609 |
| Self-defense repercussion | ||||||
| -action(s) helped the situation | 342 (70%) | 205 (64%) | 64 (80%) | 50 (74%) | 23 (82%) | 0.170 |
| -action(s) made the situation worse | 121 (32%) | 82 (33%) | 19 (33%) | 14 (26%) | 6 (28%) | 0.840 |
| Sustained injuries (yes)1 | 405 (46%) | 243 (45%) | 63 (40%) | 60 (52%) | 39 (59%) | 0.327 |
| Sustained injuries requiring medical attention2 | 167 (19%) | 93 (17%) | 33 (24%) | 26 (18%) | 15 (26%) | 0.613 |
Only asked if an attack occurred (as opposed to threat or attempted attack), N = 797
Only asked if Sustained Injuries = Yes, N = 405
Each cell displays the absolute number and weighted proportion of women in that cell who self-reported IPV to police. The sum of column percentages will not be equal to 100%.
P-value indicates differences between races for each row (e.g., among female IPV survivors who did not graduate high school, the proportion who self-reported the incident to police is significantly different for at least one of the four race/ethnicity groups). P-values were calculated using chi-square or fisher’s exact test for cell sizes less than 5.
Self-reporting IPV to Police
Forty-four percent of women self-reported their experience of IPV to police. Racial/ethnic differences emerged in reporting IPV to the police for women with less than high school education, income within $35,000-$74,999, experience of IPV-related injury, and boyfriend/girlfriend relationship type (Table 2).
Table 2.
Police reported IPV among female IPV survivors (unweighted N and weighted %), by race/ethnicity*, in the National Crime Victimization Survey
| Sample Total, N=898 | White, N=560 | Black, N=148 | Hispanic, N=126 | Other, N=64 | P-value** | |
|---|---|---|---|---|---|---|
| Reported IPV to Police | 383 (44%) | 220 (42%) | 81 (57%) | 65 (46%) | 17(35%) | 0.143 |
| Age | ||||||
| <18 years | 1 (3%) | 1 (5%) | 0 (0%) | 0 (0%) | 0 (0%) | 0.828 |
| 18–24 years | 61 (40%) | 35 (36%) | 10 (51%) | 14 (48%) | 2 (38%) | 0.713 |
| 25–34 years | 154 (51%) | 84 (53%) | 27 (71%) | 32 (43%) | 11 (38%) | 0.123 |
| 35–44 years | 97 (46%) | 56 (48%) | 27 (44%) | 13 (51%) | 1 (5%) | 0.591 |
| 45–54 years | 52 (43%) | 32 (36%) | 14 (77%) | 5 (58%) | 1 (34%) | 0.194 |
| 55–64 years | 2 (10%) | 1 (3%) | 0 (0%) | 1 (100%) | 0 (0%) | 0.535 |
| Education | ||||||
| Did not graduate high school | 55 (35%) | 24 (27%) | 17 (82%) | 11 (24%) | 3 (17%) | 0.027 |
| High school diploma | 108 (50%) | 62 (49%) | 28 (56%) | 15 (47%) | 3 (35%) | 0.735 |
| Any post-secondary education | 217 (44%) | 131 (43%) | 36 (45%) | 39 (63%) | 11 (38%) | 0.126 |
| Income | ||||||
| < $10,000 | 59 (39%) | 29 (39%) | 17 (41%) | 10 (46%) | 3 (20%) | 0.815 |
| $10,000 - $19,999 | 62 (41%) | 32 (41%) | 13 (52%) | 16 (42%) | 1 (16%) | 0.304 |
| $20,000 - $34,999 | 62 (39%) | 39 (31%) | 11 (34%) | 10 (42%) | 2 (79%) | 0.667 |
| $35,000 - $74,999 | 70 (48%) | 41 (45%) | 14 (71%) | 13 (63%) | 2 (18%) | 0.033 |
| $75,000+ | 33 (43%) | 27 (52%) | 0 (0%) | 4 (47%) | 2 (4%) | 0.210 |
| < 2015 poverty line | 114 (39%) | 60 (41%) | 31 (43%) | 19 (35%) | 4 (21%) | 0.627 |
| Living Environment | ||||||
| Urban | 333 (45%) | 178 (43%) | 79 (57%) | 64 (46%) | 12 (33%) | 0.185 |
| Rural | 50 (38%) | 42 (36%) | 2 (36%) | 1 (55%) | 5 (46%) | 0.929 |
| Public housing | 20 (33%) | 9 (22%) | 9 (64%) | 2 (25%) | - | 0.114 |
| Abuse type | ||||||
| Non-sexual | 353 (46%) | 208 (43%) | 76 (56%) | 53 (49%) | 16 (44%) | 0.343 |
| Sexual | 30 (26%) | 12 (27%) | 5 (74%) | 12 (33%) | 1 (5%) | 0.087 |
| Relationship with Offender | ||||||
| -spouse | 132 (43%) | 78 (44%) | 23 (65%) | 28 (40%) | 3 (13%) | 0.062 |
| -ex-spouse | 44 (45%) | 36 (57%) | 2 (9%) | 6 (53%) | 0 (0%) | 0.168 |
| -boy/girlfriend or ex-boy/girlfriend | 207 (44%) | 106 (37%) | 56 (63%) | 31 (49%) | 14 (43%) | 0.014 |
| Incident location | ||||||
| -Respondent’s home | 257 (45%) | 147 (45%) | 60 (58%) | 39 (44%) | 11 (28%) | 0.106 |
| -Near own home | 51 (48%) | 31 (39%) | 9 (51%) | 9 (47%) | 2 (86%) | 0.706 |
| -At or near friend/neighbor/relative’s home | 40 (37%) | 23 (30%) | 6 (60%) | 9 (53%) | 2 (40%) | 0.257 |
| -Other | 35 (37%) | 19 (40%) | 6 (47%) | 8 (50%) | 2 (8%) | 0.516 |
| Offender substance use (yes) | 156 (42%) | 87 (40%) | 33 (51%) | 23 (49%) | 13 (38%) | 0.691 |
| Weapon present (yes) | ||||||
| -any | 63 (46%) | 32 (46%) | 18 (67%) | 6 (44%) | 7 (25%) | 0.261 |
| -firearm | 20 (49%) | 11 (43%) | 5 (58%) | 2 (61%) | 2 (100%) | 0.444 |
| -sharp/edged weapon | 21 (57%) | 9 (50%) | 8 (73%) | 2 (19%) | 2 (74%) | 0.321 |
| -blunt weapon | 11 (35%) | 6 (42%) | 3 (39%) | 2 (100%) | 0 (0%) | 0.377 |
| -other weapon or type unknown | 7 (39%) | 4 (52%) | 2 (100%) | 0 (0%) | 1 (15%) | 0.326 |
| Self-defense | 244 (49%) | 145 (48%) | 56 (57%) | 36 (49%) | 7 (41%) | 0.724 |
| Self-defense outcome | ||||||
| -action(s) helped the situation | 165 (49%) | 96 (49%) | 40 (55%) | 23 (44%) | 6 (42%) | 0.835 |
| -action(s) made the situation worse | 51 (43%) | 31 (46%) | 13 (48%) | 6 (36%) | 1 (23%) | 0.829 |
| Sustained injuries | 183 (44%) | 101 (42%) | 35 (71%) | 38 (53%) | 9 (18%) | 0.008 |
| Sustained injuries requiring medical attention | 81 (43%) | 42 (39%) | 18 (73%) | 16 (53%) | 5 (15%) | 0.073 |
Each cell displays the absolute number and weighted proportion of women in that cell who self-reported IPV to police. The sum of column percentages will not be equal to 100%.
P-value indicates differences between races for each row (e.g., among female IPV survivors who did not graduate high school, the proportion who self-reported the incident to police is significantly different for at least one of the four race/ethnicity groups). P-values were calculated using chi-square or fisher’s exact test for cell sizes less than 5.
In unadjusted models, reporting IPV to police was associated with Black race (OR=1.83, 95% CI: 1.04–3.23) and increasing age between 18 and <35 years (OR=1.06, 95% CI: 1.01–1.11) (Table 3). Age of 35 years and older (OR=0.91, 95% CI: 0.85–0.98) was associated with decreasing odds of reporting; those who experienced sexual IPV versus non-sexual IPV (OR= 0.42, 95% CI: 0.19–0.94) were significantly less likely to report. In the fully adjusted model, Black women were twice as likely to report relative to White women, even after controlling for residence in public housing and abuse type (AOR=2.05, 95% CI: 1.01–4.15) (Table 3).
Table 3.
Characteristics associated with self-reporting IPV to police overall and stratified by race/ethnicity among IPV survivors, AORs (95% CI), National Crime Victimization Survey
| Unadjusted models | Full study population, N=898 | White, N=560 | Black, N=148 | Hispanic, N=126 | Other, N=64 | |
|---|---|---|---|---|---|---|
| Race/Ethnicity | ||||||
| White | ref | ref | ||||
| Black | 1.83 (1.04, 3.23) | 2.05(1.01, 4.15) | ||||
| Hispanic | 1.19 (0.70, 2.04) | 0.93(0.50, 1.74) | ||||
| Other | 0.76 (0.26, 2.21) | 1.04(0.31, 3.53) | ||||
| Aae | ||||||
| < 35 years | 1.06 (1.01, 1.11) | 1.06 (0.99, 1.14) | 1.18 (1.05, 1.33) | |||
| ≥35 years | 0.91 (0.85, 0.98) | 0.91 (0.80, 1.02) | 0.77 (0.64, 0.92) | |||
| Education | ||||||
| Did not graduate high school | 0.55 (0.29, 1.04) | 3.23 (0.98, 10.6) | 0.24 (0.07, 0.91) | |||
| High school diploma | Ref | ref | ref | |||
| Any post-secondary education | 0.81 (0.50, 1.30) | 0.64 (0.24, 1.77) | 1.77 (0.68, 4.66) | |||
| Income | ||||||
| < $10,000 | ref | ref | ||||
| $10,000 - $19,999 | 1.08 (0.54, 2.18) | 1.83 (0.38, 8.83) | ||||
| $20,000 - $34,999 | 1.00 (0.45, 2.22) | 0.80 (0.21, 3.03) | ||||
| $35,000 - $74,999 | 1.45 (0.72, 2.88) | 3.53 (0.81, 15.4) | ||||
| $75,000+ | 1.15 (0.45, 2.97) | - | ||||
| missing data | 1.49 (0.77, 2.89) | 7.33 (1.74, 31.0) | ||||
| Public housing | 0.53 (0.22, 1.25) | 0.47 (0.20, 1.09) | 0.35 (0.10, 1.16) | |||
| Abuse type | ||||||
| Non-sexual | ref | ref | ref | ref | ||
| Sexual | 0.42 (0.19, 0.94) | 0.34 (0.10, 1.19) | 0.43 (0.08, 2.31) | 0.12 (0.01, 1.44) | ||
| Relationship with offender | ||||||
| Spouse | ref | |||||
| Ex-spouse | 1.08 (0.47, 2.50) | |||||
| Current or former boy/girlfriend | 1.03 (0.63, 1.68) | |||||
| Offender substance use | 1.20 (0.69, 2.08) | |||||
| Weapon present | 1.16 (0.63, 2.15) | |||||
| Sustained injury | 0.99 (0.63, 1.54) | 2.51 (1.10, 5.71) | 2.87 (1.22, 6.71) | 0.23 (0.04, 1.41) | ||
| Injury requiring medical care | 0.96 (0.55, 1.67) |
Note: Unadjusted results give odds ratios from models which include only the specified variable and no covariates, run using the full study population. All other columns give odds ratios from the final best-fitting model for each race as specified in the top row of the table.
Stratified models identified factors associated with reporting IPV to police within racial/ethnic groups. Among Black IPV survivors, factors associated with police reporting included increasing age between 18 and <35 years (AOR=1.18, 95% CI: 1.05–1.33), IPV-related injury (AOR=2.51, 95% CI: 1.10–5.71), and declining to provide income (AOR=7.33, 95% CI: 1.74–31.0). A non-significant trend towards increased odds of reporting based on less than a high school education was also found among Black women (AOR=3.23, 95% CI: 0.98–10.6). Forty percent of Black women who did not disclose income had at least some post-secondary education. This pattern may indicate that these data are not missing at random [41, 42], and possibly signifies the socioeconomic position of women with missing data, trending towards affluence [43]. Among Hispanic women, factors related to reporting IPV to police included IPV-related injury (AOR=2.87, 95% CI: 1.22–6.71), and having less than a high school education (vs. high school graduate; AOR=0.24, 95% CI: 0.07–0.91). We did not identify statistically significant within-race determinants of reporting for White or other women.
Focus group interpretation of NCVS findings: IPV Survivors in Baltimore
Female IPV survivors who participated in focus groups were primarily 25–44 years (mean=32 years). Approximately half of the sample was Black; a majority had at least some post-secondary education (58%) and income less than $10,000 (63%). All participants were currently residing in a domestic violence shelter, and three-fourths of the women ever reported their IPV experiences to police (Table 4).
Table 4.
Demographic characteristics of focus group participants (n=19), Baltimore, Maryland
| Unweighted % (Unweighted N) | |
|---|---|
| Age (mean= 32 years) | |
| 18–24 years | 5 (1) |
| 25–34 years | 68 (13) |
| 35–44 years | 26 (5) |
| Race | |
| Black | 53 (10) |
| White | 26 (5) |
| Other | 21 (4) |
| Relationship Status | |
| Single | 74 (14) |
| Married/Separated/Divorced | 11 (2) |
| Other | 16 (3) |
| Education | |
| Did not graduate high school | 21 (4) |
| High school diploma | 21 (4) |
| Any postsecondary education | 58 (11) |
| Abuse Type | |
| Non-Sexual | 90 (17) |
| Sexual | 63 (12) |
| Reported IPV to Police | |
| Yes | 74 (14) |
| Household annual income | |
| Less than $10,000 | 63 (12) |
| $10,000 or more | 37 (7) |
Focus group results are organized by societal, community, and incident-level factors related to reporting.
Disagreement with NCVS results on Racial/Ethnic Differences and Subsequent Revelation
Initially, participants strongly disagreed with NCVS findings that Black women report IPV to police more than White women, and some questioned the legitimacy of NCVS methods (Table 5). However, through discussion, potential reasons for Black women’s greater engagement with police following IPV emerged, including IPV-related fear and severity, such as disproportionate intimate partner homicide risk. These reasons were tempered with a discussion of a lack of confidence in the justice response.
Table 5.
Triangulation Matrix of Quantitative and Qualitative Results
| NCVS Data (National) | Focus Groups (Baltimore, MD) | Level of Agreement |
|---|---|---|
| Race | ||
| Black women are more likely than White women to self-report IPV to police (adj.) | Reporting by race/ethnicity is influenced by socio-structural and incident-level factors | +/− |
| . | A lack of confidence in the justice system discourages Black women from self-reporting IPV | . |
| . | Social norms among Black women support disbanding the family unit and reporting IPV to police | . |
| Gender | ||
| . | Minimization of abuse based on gender discourages reporting to police for incidents of IPV | . |
| Age | ||
| Overall, self-report of IPV to police increases with increasing age between 18 and <35 years (unadj.) | . | . |
| Self-report of IPV to police increases with increasing age between 18 and <35 years among Black women (adj.) | . | . |
| Education | ||
| Education has no effect on reporting, overall (unadj.) | Individuals with greater education or esteemed careers (e.g., doctors, lawyers) are less likely to report IPV to police. | -- |
| Among Hispanic women, those with less than a high school education were significantly less likely to self-report IPV to police (adj.) | . | |
| Income/ Socioeconomic Status (SES) | ||
| Income has no effect on reporting, overall (unadj.) | Women with low income and fewer resources are more likely to report IPV to police than women with larger income. | -- |
| Black women who did not report their income during the survey were significantly more likely to report IPV to police (adj.) | . | . |
| . | Individuals of greater SES, needing to protect their social/professional reputation, are less likely to self-report IP to police | . |
| . | Individuals living in poverty rely on police support for IPV due to a lack of resources | |
| . | Fear of losing financial support from the IPV offender may discourage self-reporting IPV to police | . |
| Public Housing | ||
| Living in public housing has no effect on reporting (unadj) | Police presence in high crime/drug areas such as public housing facilitates passive reporting (e.g., police overhear violence). A culture of silence in urban areas impedes reporting IPV to police. | -- |
| . | Living in a high-crime neighborhood discourages self-reporting for fear of being labeled “snitch” | . |
| Abuse Type | ||
| Sexual IPV is associated with greater likelihood of self-reporting, overall (unadj.) | Survivors of sexual IPV are less likely to report it to police unless a weapon is present. | +/− |
| . | The burden of proving IPV, especially sexual IPV, discourages self-reporting to police | . |
| Relationship with Offender | ||
| Relationship with offender had no effect on reporting, overall (unadj) | . | . |
| Offender’s Substance Use | ||
| Offender’s substance use has no effect on reporting, overall (unadj) | Offender’s substance use heightened women’s fear of harm and led to reporting | -- |
| Weapon Presence | ||
| Weapon presence has no effect on reporting, overall (unadj) | Weapon presence was related to reporting IPV to police due to fear of injury and hope that the incidenct would be more believeable by police. | -- |
| Injury | ||
| Injury and injury requiring medical attention has no effect on reporting, overall (unadj) | IPV severity (injury) is related to reporting among Black women for fear of intimate partner homicide | -- |
| Among Black women, those with injury were significantly more likely to self-report IPV to police (adj.) | IPV severity (injury) is related to reporting among Black women for fear of intimate partner homicide | ++ |
| Among Hispanic women, those with injury were significantly more likely to self-report (adj.) | . | . |
| Justice System | ||
| . | Policies that allow for rapid release of offenders discourages self-reporting IPV to police | . |
| . | Victims’ fear of being arrested along with the offender discourages self-reporting IPV to police | . |
Note: Triangulation findings are based on unadjusted (unadj.) and adjusted (adj.) logistic regression models (NCVS data) and focus groups (Baltimore, MD) with recent, female IPV survivors.
Key
= agreement between the datasources
= partial agreement between the datasources
= disagreement between the datasources
= not addressed in the datasource/ unable to compare.
“I feel that Black women call the police more, probably, because it’s a fearful thing, it’s danger [IPV].”
(FG1).
“Maybe because a lot of Black women are shot.”
(FG2).
Additionally, participants explained that racial/ethnic differences in social norms regarding family structure shaped willingness to proceed as a single parent following experiences of abuse, thus increasing reporting likelihood for Black women.
[…] that’s genuinely how the Black community is raised for women. Most of the time it’s, “Forget that man. Girl you’ve got it.” […] Black women, they more so easy to be like, “What up? Call the police and put him on child support.” As opposed to a White woman. […]. A White woman will really take on a lot to try to keep her whole family as one.
(FG2)
Lack of confidence or satisfaction in police response was felt to discourage reporting, particularly for Black women.
“Us being Black females, half of which are from the inner-city East Baltimore, West Baltimore we are not running around calling police. There’s no talking to police right now because it’s not going to do anything. It’s not going to matter.”
(FG1).
Socioeconomic Status (SES) Influences Police Reporting despite Race/Ethnicity
A rich dialogue on the influence of SES emerged. Initially, participants described SES as a factor that impacts non-reporting, particularly among White women.
“They [White women] don’t want people to find out because they got reputation to keep such as lawyers, judges, people that own things and stuff.”
(FG1)
On further reflection, participants concluded that greater SES drives non-reporting regardless of race/ethnicity due to the availability of resources and social standing. Social desirability, or the desire to protect one’s professional reputation, was labeled a key reason for not engaging police in incidents of IPV among women of greater SES.
That’s not just a White woman thing [reporting]. There’s some Black women that just as bougie [bourgeoisie].
True.
Want to keep everything hush-hush. (FG1)
Women of greater SES were felt to fear social and professional ramifications of reporting, and to have the means to leave the relationship temporarily (e.g., go to a hotel) without the support of social services.
“White women don’t call the police because they scared to call. They don’t want to mess up what they got going on. They don’t want the lady down the street that work at the school to see the police at her house.”
(FG1).
Maybe it’s a financial thing too. Living in a city, maybe you have less money. I’m not saying all White people live in richer areas. I’m saying that maybe they’re less likely to call the police because they’ve got resources that they don’t need to [call police]. They can leave. They can find another residence. They can fly out of the country. They’ve got more financial security to get out.
(FG1)
“I think Black women in poverty or White women in poverty want to be protected, so they going to call the police. Black or White women with richer income, they don’t want the police in their business.”
(FG3).
Similarly, low SES, mainly fearing the loss of financial support, was described as a barrier to reporting.
“There was that chance I was taking [by calling police] that everything was going to change--knowing if he gets taken off, that’s one less person helping me. What do I do? How do I afford rent on my own?”
(FG2).
Participant reflection on the intersectionality of SES and race/ethnicity prompted an exploration of income and missingness in the quantitative analysis.
Discrimination by Gender and Race Discourages Reporting
Past discrimination by gender and race undermined participants’ confidence to engage with police. Participants’ experiences with law enforcement varied based on responding officer’s gender and were perceived to be influenced by uneven power dynamics and gender norms for incidents of gender-based violence.
“When it’s a man [police officer], they’re strictly business. ‘Is it time to go yet? It’s about time for his lunch break.’ When it’s a woman, she’s compassionate. ‘I’m so sorry this happened to you,’ or ‘Are you OK? Let’s get you to the hospital.’ that sort of stuff.”
(FG3).
Minimization of abuse as a deterrent to reporting was a common theme in all focus groups. Participants described being burdened by a lack of proof that abuse occurred, especially for incidents of sexual IPV.
It’s hard for people to believe that you were getting raped inside of a consensual relationship. For some reason, men, I don’t know if it’s men or it’s just being in the position of power when you’re a cop, but they already have answers. ‘Well, that’s probably not what happened. Are you sure that that’s what happened?’
(FG1).
In addition to perceived gender discrimination, Black women described racism and classism as barriers to reporting.
“Personally, I don’t do the police because it’s not going to matter. […]. They already got in their mind, ‘She’s nothing but another n*gger that’s going through whatever it is that she’s going through. They’re going to look at us with a stereotype.”
(FG1)
Mistrust in the IPV Justice Response
For some, barriers to self-reporting IPV extended beyond prior engagement with police to a general lack of confidence in the justice process for IPV.
“I’m looking at the police but I’m thinking judicial. It’s not just the police that I need help from. It’s the police’s job to arrest him but it’s not his job to keep him there.”
(FG2)
“This is his 17th arrest for the same thing. They’ll get out and they’ll just do it all like they don’t think they did anything wrong.” (FG3)
“Just walk out [of jail] and just get overnight stay or whatever. They [IPV offender] come home and they are outraged because you ruined their night, especially if they were drinking or partying.”
(FG3)
Some participants feared that they, too, would be arrested for the IPV incident.
“A lot of people don’t report to the police, there’s consequences when you report. You [the victim] might go to jail. […] Yeah, both might go to jail. […] usually they take everybody.”
(FG3).
Community Context and Cultural Norms Influence Reporting
Police presence in high crime areas indirectly facilitates reporting.
“Police are more around certain high-drug areas, so they’ll hear or see you outside and pull over.”
(FG3)
Some IPV victims feared retribution from fellow community members or being labeled a “snitch” for engaging with police due to a generalized culture of silence in response to crime.
Us calling the police in a high-crime area, just them (neighbors) seeing the police come to our house and then talking to them about anything, we going to be labeled “snitches,” which means we’ll have problems in our neighborhood. If something happens and the police be called, they going to think it’s us, because we called the police [before].
(FG2)
Weapon Presence and Severe IPV increase Self-reporting
In responding to infographics related to differences in police reporting based on IPV severity and weapon presence, participants explained that severe IPV and incidents with weapons present were “more believable” and harder to control without police.
“Police might do more considering [weapon presence]. They don’t know if the person is doing something sexually violent to you. Using a weapon might make it more believable for the police to actually come and do something.
(FG2)
“There was that absolute fear that I can’t control this, he won’t listen, he had been drinking. […]. I didn’t have control over the situation and there was a chance that I could die. That was when I called the cops.”
(FG2)
The strength of our sequential mixed-methods approach is evidenced by our triangulation analysis, which demonstrates the level of complementarity and uniqueness of data (Table 5). Triangulation of results across data sources (NCVS and local focus groups) identified agreement on the influence of IPV severity on police reporting for Black women. Partial disagreement was identified regarding the influence of race/ethnicity on police reporting. Areas of disagreement between the data sources include education, income, and weapon presence during an IPV incident. Varied areas of silence, indicated by a dot in the triangulation matrix, demonstrate the benefit of synergizing data sources(Table 5).
DISCUSSION
Self-reporting IPV to the police varies significantly by race/ethnicity due to varied influences of age, income, education, and injury severity. This study is a first step to understanding police reporting among IPV survivors through an intersectional lens of race and gender, using national estimates from the NCVS. Furthermore, focus groups with IPV survivors in Baltimore about key NCVS findings contextualize when and how particular factors influence reporting and for whom, focusing on social and structural factors omitted from the NCVS.
Our stratified approach suggests racial/ethnic differences in drivers of police reporting that have been missed in traditional analyses that simply control for race/ethnicity. These findings emphasize the varied impact of injury severity, age, education, and income by race/ethnicity. In the full model, Black women were twice as likely to report IPV as White women, which is inconsistent with strongly-held theories that Black people do not engage police. Black and Hispanic men and women engage police less [18, 44, 45], but this pattern is untrue for IPV incidents [9, 10].
Subsequent stratified models underscore severe IPV-injury as a driver of police reporting among Black and Hispanic women relative to uninjured women of the same race/ethnicity. Younger age was also a significant predictor of reporting among Black women only. Age-range 18 to 24 is a particularly vulnerable period for homicides among women, especially for Black, Hispanic, and Pacific Island women [2]. Focus group participants conclude that perhaps reporting is driven by economic rather than racial/ethnic factors, as women with greater financial means can flee abuse without supportive services, and individuals of greater SES receive more respect from law enforcement and the greater society. Participants also discussed the disproportionate risk of intimate partner homicide among Black women [2, 5] and how fear of lethality may serve as the catalyst for reporting IPV to police. The influence of having less than high school education, often a proxy for economic instability, on reporting trended in opposite directions for Black and Hispanic women, hindering reporting among Hispanic women. In short, participatory focus group discussions extend our understanding of the influence of underlying racial tension, systemic minimization of gender-based violence, SES, and community norms on reporting, and identify significant gaps in the available NCVS indicators for understanding police contact patterns. Together, these findings demonstrate the nuance of race/ethnicity on IPV reporting.
Fear of deportation and linguistic barriers may explain underreporting among Hispanic women [46]. These barriers to police engagement are supported in our racial/ethnic comparison. Education had a reverse effect on reporting among Hispanic women in the NCVS relative to Black women. Discrimination from police based on citizenship, and social norms that detest divorce were discussed as possible reasons for the varied directional impact of education on reporting. Participants’ discussion of SES prompted subsequent analysis of the NCVS income variable.
Our findings clarify findings from previous studies on the influence of SES and neighborhood poverty on police reporting for IPV among racial/ethnic groups [10, 46]. Income did not affect reporting in our full, adjusted model of NCVS data. However, the influence of education and the amount of missing income data among Black women who reported IPV to police supports the impact of SES. Non-report of income may dovetail with social desirability described by focus group participants, which includes influence from social norms and SES. While it is intuitive that individuals with fewer resources may be more likely to rely on supportive services, our findings underscore the impact of social barriers that exist. Among urban, crime-ridden communities, fear of being labeled a police informant impedes reporting. Similarly, individuals with esteemed occupations may be hesitant to contact the police due to IPV-related stigma and fear that others may learn of the abuse. Future studies should explore the influences of race/ethnicity on police reporting among a sample of high-income women and where and how they may seek additional support.
Police are gateways to victim support and safety, but community norms may delay such engagement in the absence of severe, potentially fatal abuse [47–49]. In our study and others, police reporting was significantly associated with IPV severity among Black and Hispanic women [47, 48]. However, this finding was conflated by the discussion of self-defense implications as an impeding factor of reporting among focus group participants. Victims of abuse, particularly in marginalized communities, may be reluctant to report IPV to police for fear of being arrested themselves [50]. Escalated forms of abuse predict IPV homicide [11], with disproportionate risk among Black women [2, 5]. In a recent study of female intimate partner homicide victims, 91% had some engagement with police within the three years leading up to their death [30]. Mitigating barriers to self-reporting IPV to police and finding ways to repair community trust in law enforcement are necessary next steps for intervention. In addition, our findings and others point to the need for guided, intermediate interventions, possibly outside of the justice system, that prioritize victims’ safety and their preferences for safety planning.
Limitations
The NCVS data are limited regarding abuse frequency and socio-structural risk factors. Income effects should be interpreted with caution due to potential reporting bias, particularly for Black women. Small cell sizes influenced the extent of racial/ethnic comparisons. Qualitative activities were conducted in a single city; their broader generalizability is unknown.
Implications
Our study capitalizes on the racial/ethnic diversity of the NCVS sample and extends interpretation of results by including perspectives of IPV victims and triangulation. Together, findings highlight the influence of social and structural forces and the need for a more in-depth and inclusive study of IPV victims’ police reporting practices to inform future policy and practice.
Our stratified analyses contribute to the identification of risk factors within groups, overcoming the overgeneralization of health outcomes and associated determinants from analyses that do not sufficiently examine racial/ethnic differences. Future research should clarify and extend our understanding of racial/ethnic differences in police reporting for IPV; this work will require quantification of the social and structural influences identified through qualitative research yet unmeasured in the NCVS. Our study adds to a growing body of work that demonstrates an urgent need to mend community-police relationships, particularly for communities of color. Overcoming IPV reporting barriers has implications for enhancing survivor health, safety, and well-being [49], and is a next-step in improving law enforcement’s response to IPV as a public health issue [51].
Acknowledgments
The authors acknowledge IPV/SV advocates and service providers in Baltimore, MD for their support and the survivors’ whose stories are captured in this article.
Funding
This study was supported with funding from the Health Resources and Services Administration, Maternal and Child Health Bureau (T76MC00003), Bloomberg American Health Initiative (Spark Award, Decker), National Institute of Minority Health and Health Disparities (1L60MD012089-01, Holliday; 5U54MD000214-17), and National Institute of Mental Health (1T32MH109436-01A1).
Footnotes
Publisher's Disclaimer: This Author Accepted Manuscript is a PDF file of a an unedited peer-reviewed manuscript that has been accepted for publication but has not been copyedited or corrected. The official version of record that is published in the journal is kept up to date and so may therefore differ from this version.
Author Disclosure Statement
No competing financial interests exist.
Informed consent
All participants provided informed consent.
Ethics Approval
The Johns Hopkins Bloomberg School of Public Health’s IRB approved all study procedures.
REFERENCES
- 1.Smith SG, Zhang X, Basile KC, Merrick MT, Wang J, Kresnow MJ et al. The National Intimate Partner and Sexual Violence Survey: 2015 Data Brief – Updated Release. National Center for Injury Prevention and Control, Centers for Disease Control and Prevention, Atlanta, GA: 2018. https://www.cdc.gov/violenceprevention/pdf/2015data-brief508.pdf. Accessed 10/15/2019. [Google Scholar]
- 2.Petrosky E, Blair JM, Betz CJ, Fowler KA, Jack SPD, Lyons BH. Racial and Ethnic Differences in Homicides of Adult Women and the Role of Intimate Partner Violence - United States, 2003–2014. Mmwr-Morbid Mortal W. 2017;66(28):741–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Baker CK, Billhardt KA, Warren J, Rollins C, Glass NE. Domestic violence, housing instability, and homelessness: A review of housing policies and program practices for meeting the needs of survivors. Aggression and Violent Behavior. 2010;15(6):430–9. doi: 10.1016/j.avb.2010.07.005. [DOI] [Google Scholar]
- 4.Campbell JC. Health consequences of intimate partner violence. The Lancet. 2002;359(9314):1331–6. doi: 10.1016/S0140-6736(02)08336-8. [DOI] [PubMed] [Google Scholar]
- 5.When Men Murder Women An Analysis of 2016 Homicide Data. Washington, DC: Violence Policy Center; 2018. [Google Scholar]
- 6.Srinivas T, DePrince AP. Links Between the Police Response and Women’s Psychological Outcomes Following Intimate Partner Violence. Violence and victims. 2015;30(1):32–48. doi: 10.1891/0886-6708.Vv-d-13-00109. [DOI] [PubMed] [Google Scholar]
- 7.Hanson GC, Messing JT, Anderson JC, Thaller J, Perrin NA, Glass NE. Patterns and Usefulness of Safety Behaviors Among Community-Based Women Survivors of Intimate Partner Violence. J Interpers Violence. 2019:886260519853401. doi: 10.1177/0886260519853401. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Decker MR, Holliday CN, Hameeduddin Z, Shah R, Miller J, Dantzler J et al. “You don’t think of me as a human being”: Race and gender inequities intersect to discourage police reporting of violence against women. Journal of Urban Health. 2019. doi: 10.1007/s11524-019-00359-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Chen Y, Ullman SE. Women’s reporting of sexual and physical assaults to police in the National Violence Against Women Survey. Violence against women. 2010;16(3):262–79. doi: 10.1177/1077801209360861. [DOI] [PubMed] [Google Scholar]
- 10.Ackerman J, Love TP. Ethnic group differences in police notification about intimate partner violence. Violence against women. 2014;20(2):162–85. doi: 10.1177/1077801214521327. [DOI] [PubMed] [Google Scholar]
- 11.Campbell JC, Webster D, Koziol-McLain J, Block C, Campbell D, Curry MA et al. Risk factors for femicide in abusive relationships: results from a multisite case control study. American journal of public health. 2003;93(7):1089–97. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Messing JT, Campbell J, Sullivan Wilson J, Brown S, Patchell B. The Lethality Screen: The Predictive Validity of an Intimate Partner Violence Risk Assessment for Use by First Responders. J Interpers Violence. 2017;32(2):205–26. doi: 10.1177/0886260515585540. [DOI] [PubMed] [Google Scholar]
- 13.Zeoli AM, McCourt A, Buggs S, Frattaroli S, Lilley D, Webster DW. Analysis of the Strength of Legal Firearms Restrictions for Perpetrators of Domestic Violence and Their Associations With Intimate Partner Homicide. Am J Epidemiol. 2018;187(11):2365–71. doi: 10.1093/aje/kwy174. [DOI] [PubMed] [Google Scholar]
- 14.Lipsky S, Caetano R, Roy-Byrne P. Racial and ethnic disparities in police-reported intimate partner violence and risk of hospitalization among women. Womens Health Issues. 2009;19(2):109–18. doi: 10.1016/j.whi.2008.09.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Smith Stover C. Police–Advocacy Partnerships in Response to Domestic Violence. Journal of Police Crisis Negotiations. 2012;12(2):183–98. doi: 10.1080/15332586.2012.717031. [DOI] [Google Scholar]
- 16.Dichter ME, Gelles RJ. Women’s perceptions of safety and risk following police intervention for intimate partner violence. Violence against women. 2012;18(1):44–63. doi: 10.1177/1077801212437016. [DOI] [PubMed] [Google Scholar]
- 17.Truman JL, Morgan RE. Criminal Victimization, 2015. U.S. Department of Justice, Bureau of Justice Statistics. 2016. https://www.bjs.gov/content/pub/pdf/cv15.pdf. Accessed NCJ 250180.
- 18.National Public Radio, Robert Wood Johnson Foundation, Harvard T.H. Chan School of Public Health. Discrimination in America: Experiences and Views of African Americans. 2017.
- 19.Washington PA. Disclosure Patterns of Black Female Sexual Assault Survivors. Violence against women. 2001;7(11):1254–83. doi:doi: 10.1177/10778010122183856. [DOI] [Google Scholar]
- 20.Langton L, Berzofsky M, Krebs C, Smiley-McDonald H. National Crime Victimization Survey: Victimizations Not Reported To Police, 2006–2010. Washington, DC: Bureau of Justice Statistics, Office of Justice Programs, US Department of Justice; 2012. [Google Scholar]
- 21.Marchetti CA. Regret and Police Reporting Among Individuals Who Have Experienced Sexual Assault. Journal of the American Psychiatric Nurses Association. 2012;18(1):32–9. doi: 10.1177/1078390311431889. [DOI] [PubMed] [Google Scholar]
- 22.Fedina L, Backes BL, Jun HJ, Shah R, Nam B, Link BG et al. Police violence among women in four US cities. Prev Med. 2018;106:150–6. doi: 10.1016/j.ypmed.2017.10.037. [DOI] [PubMed] [Google Scholar]
- 23.Decker MR, Pearson E, Illangasekare SL, Clark E, Sherman SG. Violence against women in sex work and HIV risk implications differ qualitatively by perpetrator. BMC public health. 2013;13:876. doi: 10.1186/1471-2458-13-876. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Holliday CN, Morse SM, Irvin NA, Green-Manning A, Nitsch LM, Burke JG et al. Concept Mapping: Engaging Urban Men to Understand Community Influences on Partner Violence Perpetration. Journal of urban health : bulletin of the New York Academy of Medicine. 2018;96(1):97–111. doi: 10.1007/s11524-018-0297-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Archibald PC, Parker L, Thorpe R Jr, Criminal Justice Contact, Stressors, and Obesity-Related Health Problems Among Black Adults in the USA. J Racial Ethn Health Disparities. 2018;5(2):387–97. doi: 10.1007/s40615-017-0382-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Bor J, Venkataramani AS, Williams DR, Tsai AC. Police killings and their spillover effects on the mental health of black Americans: a population-based, quasi-experimental study. The Lancet. 2018;392(10144):302–10. doi: 10.1016/s0140-6736(18)31130-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Sherman SG, Footer K, Illangasekare S, Clark E, Pearson E, Decker MR. “What makes you think you have special privileges because you are a police officer?” A qualitative exploration of police’s role in the risk environment of female sex workers. AIDS Care. 2015;27(4):473–80. doi: 10.1080/09540121.2014.970504. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Alang S, McAlpine D, McCreedy E, Hardeman R. Police Brutality and Black Health: Setting the Agenda for Public Health Scholars. American journal of public health. 2017;107(5):662–5. doi: 10.2105/AJPH.2017.303691. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Edwards F, Esposito MH, Lee H. Risk of Police-Involved Death by Race/Ethnicity and Place, United States, 2012–2018. American journal of public health. 2018;108(9):1241–8. doi: 10.2105/ajph.2018.304559. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Koppa V, Messing JT. Can Justice System Interventions Prevent Intimate Partner Homicide? An Analysis of Rates of Help Seeking Prior to Fatality. J Interpers Violence. 2019:886260519851179. doi: 10.1177/0886260519851179. [DOI] [PubMed] [Google Scholar]
- 31.Gomez MB. Policing, Community Fragmentation, and Public Health: Observations from Baltimore. Journal of urban health : bulletin of the New York Academy of Medicine. 2016;93 Suppl 1:154–67. doi: 10.1007/s11524-015-0022-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.US Department of Justice. Investigation of the Baltimore City Police Department. Washington, DC: U.S. Department of Justice, Civil Rights Division; 2016. [Google Scholar]
- 33.Survey Methodology for Criminal Victimization in the United States. 2019.
- 34.U.S. Department of Justice BoJS. Data Collection: National Crime Victimization Survey (NCVS). 2018. https://www.bjs.gov/index.cfm?ty=dcdetail&iid=245. Accessed 10/16/2019. [Google Scholar]
- 35.Holliday CN, McCauley HL, Silverman JG, Ricci E, Decker MR, Tancredi DJ et al. Racial/Ethnic Differences in Women’s Experiences of Reproductive Coercion, Intimate Partner Violence, and Unintended Pregnancy. Journal of women’s health (2002). 2017;26(8):828–35. doi: 10.1089/jwh.2016.5996. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36.RTI International. SUDAAN 11. Research Triangle Park, NC2012. [Google Scholar]
- 37.SAS Institute I. SAS 9.4. Cary, NC: SAS Institute, Inc.; 2013. [Google Scholar]
- 38.Kane M, Trochim WMK. Concept mapping for planning and evaluation Applied social research methods series, vol 50. Thousand Oaks: Sage Publications; 2007. [Google Scholar]
- 39.LeCompte M, Schensul J. Analyizing and Interpreting Ethnographic Data Ethnographers’ ToolKit. Walnut Creek, PA: SAGE Press; 1999. [Google Scholar]
- 40.Farmer T, Robinson K, Elliott SJ, Eyles J. Developing and implementing a triangulation protocol for qualitative health research. Qual Health Res. 2006;16(3):377–94. doi: 10.1177/1049732305285708. [DOI] [PubMed] [Google Scholar]
- 41.Chen JT, Kaddour A, Krieger N. Implications of missing income data. Public Health Rep. 2008;123(3):260; author reply −1. doi: 10.1177/003335490812300303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 42.Kim S, Egerter S, Cubbin C, Takahashi ER, Braveman P. Potential implications of missing income data in population-based surveys: an example from a postpartum survey in California. Public Health Rep. 2007;122(6):753–63. doi: 10.1177/003335490712200607. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43.Braveman P, Cubbin C, Marchi K, Egerter S, Chavez G. Measuring socioeconomic status/position in studies of racial/ethnic disparities: maternal and infant health. Public Health Rep. 2001;116(5):449–63. doi: 10.1093/phr/116.5.449. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 44.Durose M, Langton L. Requests for police assistance, 2011. U.S. Department of Justice, Bureau of Statistics,. 2013. https://www.bjs.gov/content/pub/pdf/rpa11.pdf. [Google Scholar]
- 45.Carr PJ, Napolitano L, Keating J. We never call the cops and here is why: A qualitative examination of legal cynicism in three philadelphia neighborhoods. Criminology. 2007;45(2):445–80. doi:DOI 10.1111/j.1745-9125.2007.00084.x. [DOI] [Google Scholar]
- 46.Pearlman DN, Zierler S, Gjelsvik A, Verhoek-Oftedahl W. Neighborhood environment, racial position, and risk of police-reported domestic violence: a contextual analysis. Public Health Rep. 2003;118(1):44–58. doi: 10.1093/phr/118.1.44. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 47.Carbone-López KC. The ‘Usual Suspects’: How Race Affects Decisions to Report Rape Victimization. Journal of Ethnicity in Criminal Justice. 2005;3(4). [Google Scholar]
- 48.Hyman I, Forte T, Du Mont J, Romans S, Cohen MM. Help-seeking behavior for intimate partner violence among racial minority women in Canada. Womens Health Issues. 2009;19(2):101–8. doi: 10.1016/j.whi.2008.10.002. [DOI] [PubMed] [Google Scholar]
- 49.Xie M, Lynch JP. The Effects of Arrest, Reporting to the Police, and Victim Services on Intimate Partner Violence. Journal of Research in Crime and Delinquency. 2017;54(3):338–78. doi: 10.1177/0022427816678035. [DOI] [Google Scholar]
- 50.McCormack PD, Hirschel D. Race and the Likelihood of Intimate Partner Violence Arrest and Dual Arrest. Race and Justice. 2018. doi: 10.1177/2153368718802352. [DOI] [Google Scholar]
- 51.Addressing Law Enforcement Violence as a Public Health Issue, Policy Number 201811. 2018. https://www.apha.org/policies-and-advocacy/public-health-policy-statements/policy-database/2019/01/29/law-enforcement-violence.