Abstract
This study involves an evaluation of an innovative approach to the handling of domestic violence (DV) cases in the city of Cleveland, Ohio that includes (1) a DV Project composed of specially trained police detectives, prosecutors and victim advocates for investigating and prosecuting domestic violence cases involving adult female victims who are married to, cohabitating with, or have a child with the defendant; and (2) a Dedicated Domestic Violence Docket that involves two Municipal Court judges hearing all of the domestic violence cases that are handled by the DV Project. We collected data on six months of domestic violence cases occurring in the latter half of 2008 (N = 1388), by linking records from the Cleveland Police Department, the Prosecutor’s Office, and the Municipal Court. We found that very few victims in police districts lacking the DV Project follow up with a prosecutor to pursue the case further, indicting that specialized DV units in police departments can have a significant impact on the number of DV cases that move forward through the criminal justice system. DV Project cases were slightly less likely to result in charges issued by prosecutors (OR = .499) but more likely to result in dismissals (OR = 2.545) and referrals to DV treatment programs (χ2 = 3.88).
Similar content being viewed by others
According to the Bureau of Justice Statistics, 22% of female non-fatal violent victimizations and 30% of female homicides are committed at the hands of a male intimate or ex intimate partner (Catalano, 2007; Lauritsen, Owens, Planty, Rand, & Truman, 2012). Given the likelihood that self-report data underestimate the true prevalence of intimate partner violence, the actual figures may be as high as one in three women experiencing a physical assault from a male partner during adulthood (Black et al., 2011; Browne, 1997). The bright spot in these otherwise dismal statistics is that intimate partner domestic violence incidents have declined since 1994 (Catalano, 2012). However, they still represent a large portion of non-fatal violent victimization, accounting for more than 800,000 reported victimizations annually (Truman & Morgan, 2016). Furthermore, reported acts of minor violence appear to have increased substantially (Stark, 2007).
Several possible explanations exist for the decrease in domestic violence incidents. The most empirically supported explanation emphasizes changes in the availability of resources to battered women, including domestic violence hotlines, shelters, and legal advocacy (Dugan, Nagin, & Rosenfeld, 1999, 2003). Modifications in public perceptions of domestic violence are another possibility (Buzawa & Buzawa, 2008). Changes occurring in the criminal justice response to the problem have also been examined as factors contributing to the decline. These changes have been most pronounced in the area of police use of mandatory arrest policies for domestic violence. During the 1980s, mandatory and pro-arrest policies in these types of cases became common in police departments (Jones & Belknap, 1999). While research has provided mixed reviews regarding whether these mandatory arrest policies lessen recidivism (Davis, O’Sullivan, Farole Jr., & Remple, 2008; Iyengar, 2009; Orchowsky, 1999; Xie, Lauritsen, & Heimer, 2012), police departments across the United States have widely implemented mandatory or pro-arrest policies in domestic violence cases (Buzawa, Buzawa, & Stark, 2012; Xie et al., 2012).
One result of these mandatory arrest policies is an increase in these types of cases seen by courts (Dawson & Dinovitzer, 2001; Friday, Lord, Exum, & Hartman, 2006). These cases present a number of unique problems for the criminal justice system and are often a source of frustration for police and prosecutors. For example, one of the most significant impediments to prosecuting defendants in domestic violence cases is victims are often unwilling to cooperate (Belknap, Hartman, & Lippen, 2010) and/or change their minds about prosecution due to factors such as fear of their abuser, economic dependency on the offender, or having children in common (Davis, Smith, & Nickles, 1997; Sigler, Crowley, & Johnson, 1990). As a result, prosecutors often find it necessary to drop the case as a result of the victim’s unwillingness to testify. Thus, in addition to using mandatory arrest policies some jurisdictions have also implemented specialized prosecution strategies aimed at prosecuting cases without the victims’ consent (Dawson & Dinovitzer, 2001).
Domestic violence cases also present unique complexities with respect to overall judicial handling given the nature of the victim/offender relationship and the relatively high likelihood the relationship between the victim and defendant will continue and/or resume after initial contact with the criminal justice system. Victims are often fearful of retaliation from the defendant for cooperating with criminal justice agencies and may be subject to threats and intimidation throughout the process (Bennett, Goodman, & Dutton, 1999; Erez & Belknap, 1998; Singer, 1988). Thus, specialized police and prosecution units as well as specialized courts have been developed to combat the various obstacles associated with these types of cases. The purpose of this paper is to present the findings of an evaluation of specialized units within Cleveland Ohio’s police department, prosecutor’s office, and municipal court for processing domestic violence cases. After describing the creation of these units, we discuss the quasi-experimental design of the study, followed by our findings and their policy implications.
Specialized Domestic Violence Police Units
Although mandatory arrest policies are the most widely discussed law enforcement strategy for addressing domestic violence offenses, specialized police domestic violence units have emerged as another approach (Friday et al., 2006). According to a national survey of 14,000 law enforcement agencies, 11% reported having a specialized domestic violence unit (Townsend, Hunt, Kuck, & Baxter, 2005). The percent (56) was much higher in larger departments (those having 100 or more officers). Diversity exists in the function of these units, which can include the appointment of specially trained detectives to investigate the cases, victim assistance with securing protective orders, victim safety planning, and referrals to counseling.
Specialized Domestic Violence Prosecution Units
The development of specialized units within prosecutors’ offices across the country has been another tactic used to reduce dismissal rates and increase victim cooperation in domestic violence cases. Specialized prosecution programs encompass a variety of strategies to accomplish these goals, including the adoption of no-drop policies, specialized domestic violence training for prosecutors, and vertical prosecution strategies. Another goal of specialized prosecution programs has been the promotion of evidence based (sometimes called “victimless”) prosecution in domestic violence cases (Davis et al., 2008). This type of prosecution relies heavily on physical evidence and third party testimony instead of victim testimony (Dawson & Dinovitzer, 2001).
Specialized Domestic Violence Courts
Beginning in the 1980s, a number of different types of problem-solving courts emerged across the country, including those focused on offenders with drug and mental health problems. In contrast to these other forms of specialized courts, the development in the early 1990s of domestic violence courts sought to ensure the safety of the victim while also addressing the actions of the offender (Buzawa et al., 2012; Labriola, Bradley, O’Sullivan, Rempel, & Moore, 2009). The growth of domestic violence courts was further aided by the passage of the Violence Against Women Act in 1994, which provided funding to increase the adjudication of individuals committing violent offenses against women (Shelton, 2007). While some specialized courts handle family violence cases, many focus exclusively on intimate partner violence, as does the court examined in the current study. The main focus of these domestic violence courts currently is on making offenders accountable for their behavior (Peterson, 2008; Shelton, 2007).
Innovative criminal justice approaches such as these aiming to improve the handling of domestic violence cases have been around sufficiently long that a growing body of literature exists seeking to evaluate them. The methodological approaches used in these evaluations vary considerably, as does the extent to which the studies meet rigorous, scientific standards for program evaluation. A major obstacle in these evaluations has been the lack of appropriate comparison data, making it difficult to fully assess the impacts of these approaches. Shortcomings aside, the outcomes of these evaluations on the whole are not overwhelmingly positive (Garcia & McManimon, 2011; Robinson & Maxwell, 2008; Spohn, 2008). In part, this may reflect the difficulty of developing strategies to simultaneously achieve two competing goals: increased offender accountability and increased protection of victims (Johnson & Dawson, 2011).
Cleveland’s Domestic Violence Project
Beginning in the late 1990s, three major players in Cleveland’s criminal justice system (the police, prosecutors, and victim advocates) sought ways to improve the handling of domestic violence cases in the city. Along with a more recent initiative by Cleveland’s Municipal Court, these efforts have collectively come to be known as the “Domestic Violence Project” (see Fig. 1). The initiatives began in 1997 with funding provided by a Violence Against Women Act STOP Block Grant under the federal Violence Against Women Act, which pays for a portion of the DV Project detectives’, prosecutors’, and victim advocates’ salaries. The stated objectives of this approach to handling domestic violence cases in the City of Cleveland as identified in their initial requests for funding were an increased number of investigations, charges, and successful prosecutions in domestic violence cases. The final aspect of the project, Cleveland’s Dedicated Domestic Violence Docket, was created by Municipal Court Judge Ronald Adrine. The Dedicated Domestic Violence Docket began taking domestic violence cases in January 2007.
Components of Cleveland’s DV Project
Cleveland’s Police Department contains a specialized domestic violence unit. At the time we conducted this study, there were eight detectives assigned to this unit. In accordance with the Violence Against Women Act grant, the unit handles cases of domestic violence in which the victim is an adult female and the victim and defendant are either married, cohabitating, or have a child together. Detectives are provided reports of domestic violence offenses from the second, third, and, fourth districts of the city and further investigate the cases by gathering additional evidence and attempting to contact the victim. Second, the Cuyahoga County Witness/Victim Center has approximately eight victim advocates assigned to the Domestic Violence Project. Advocates also seek to make contact with the victim and work with them to apply for civil protection orders, create a safety plan, and accompany them to court. Third, the City of Cleveland’s Prosecutor’s Office has three prosecutors and a law clerk assigned to handle domestic violence cases. The Prosecutors review cases presented by the detectives, issue charges (both misdemeanor and felony), and prosecute misdemeanor DV cases. Finally, cases investigated by the police department’s DV unit and charged by one of the DV prosecutors as misdemeanors are then assigned to the Dedicated Domestic Violence Docket by the Clerk’s Office. Initially two of the thirteen Municipal Court judges heard all of the cases assigned to the Dedicated Domestic Violence Docket. That number has since been increased to three judges.
The Cleveland DV Project adheres to many of the objectives identified in the literature on specialized units within the criminal justice system. Evidence-based prosecution is emphasized. However, if both evidence and victim cooperation are lacking, the prosecutor may decline to issue charges. Cases in which charges are issued continue to be handled by the DV prosecutor and are assigned to one of the judges on the Dedicated Domestic Violence Docket, who have attended specialized training in domestic violence issues. These same judges also hold compliance hearings for any domestic violence offenders they convict and sentence to probation.
The Domestic Violence Project currently handles cases where the incident occurred in Cleveland’s second, third, and fourth police districts. No DV Project exists in the remaining two police districts in the city (districts one and five). The goal has always been to further expand the Project to encompass all five districts. This goal has not been met for several reasons, including financial barriers and limits to the number of city prosecutors that can be hired.
In terms of demographics, the non-DV Project districts are somewhat higher in terms of socioeconomic status than the DV Project districts. For example, the average unemployment rate across the DV Project districts is 19.8% while in the non-DV Project districts it is 16.9%.Footnote 1 The poverty rate is approximately 36% in the DV Project districts and 28% in the non-DV Project districts. The non-DV project districts also have proportionately fewer African American residents (49.3% compared to 58.4% for the DV Project districts). The age distribution of the population in the DV Project and non-DV Projects is similar. Approximately 24% of the population in each is under 18 years of age and 12% is 65 years and older. The distribution of 18 to 29 year olds is 20.7% and 17.6%, 30 to 49 year olds is 25.1% and 27.1%, and 50 to 64 year olds is 18.1% and 19.4% in the DV Project and non-DV Project districts respectively. The rates of domestic violence in the DV Project and non-DV Project districts for the period under investigation (July to December 2008) are also similar: 6.72 and 6.32 per 1000 respectively.
Methodology
Research Design
The delay in expanding the DV Project citywide to encompass all five districts provided a unique opportunity to evaluate this new strategy for handling domestic violence cases by allowing for a natural control group. In particular, it allowed for the possibility of comparing cases handled by the police department’s DV detectives, the Cleveland DV Prosecutors, and the Municipal Court’s Dedicated Domestic Violence Docket to those cases that continued to be handled by regular district detectives and any of the remaining judges that hear criminal cases on the Municipal Court. Thus, this study employed a quasi-experimental design using an experimental and control group. We compared the outcomes of cases handled by the Domestic Violence Project (the experimental group) with those handled in the districts currently not part of the project (control group). For cases in the control group, if the DV incident was a misdemeanor, the victim was given a referral to meet with a City Prosecutor (requiring the victim to go the Prosecutor’s office downtown). If the DV incident was a felony, the case would be assigned to the district detective bureau to be handled by one of their investigators. Control group cases were not reviewed or handled by a specialized DV prosecutor or assigned to the Dedicated Domestic Violence Docket.
By linking data collected from the Cleveland Police Department’s files on domestic violence cases, the Prosecutor’s Office, and the Cleveland Municipal Court, we tracked each domestic violence case that fit the Violence Against Women Act grant criteria (see above) for the entire city for a 6-month period. Only cases originating through the police department were included. For the DV Project districts, we used the logbook in the Cleveland Police Department’s Domestic Violence Unit to identify all of the domestic violence reports from those districts meeting the Violence Against Women Act criteria. All reports sent down to the Domestic Violence Unit are reviewed against the Violence Against Women Act criteria; those that do not meet it are returned to the districts. For the non-DV Project districts we had to go out to each district and read through all of the domestic violence reports for the period under investigation to determine which met the Violence Against Women Act criteria. The overall percent of domestic violence cases that met the Violence Against Women Act criteria across all districts was 52.5%; however, this percent ranges from 29.5% to 62.6% across the districts.
Measures
With the exception of the court information, data were collected from paper files. Following other research examining the impacts of specialized domestic violence courts (Dawson & Dinovitzer, 2001; Newmark, Rempel, Diffily, & Kane, 2001), we collected basic demographic information on the victim and defendant, the victim/offender relationship, time and location of the offense, prior criminal history of the defendant, the nature of the charges laid, whether or not the case went to trial, existence and type of conviction, and sentencing information.Footnote 2 On the basis of these data, we were able to compare case aspects such as the rate of completed prosecution, the rate of case dismissals and guilty findings, type and length of sentence, and recidivism rates for cases handled through the Domestic Violence Project versus those handled by the non-participating police districts.
Analysis
Significance tests were conducted using Chi-Square (for categorical variables) and t-tests (for continuous variables). Because we identified significant differences in the racial and ethnic make-up of the suspects and victims across the five districts, we included controls for suspect and victim race in our multivariate models predicting charges and dismissals in the form of a dummy variable for black or non-black (where the largest differences across districts were evident).
We conducted a more qualitative assessment of the impact of the Dedicated Domestic Violence Docket in the form of a process evaluation by organizing focus groups and interviews with detectives in the DV Unit of the Cleveland Police Department, the DV Prosecutors in the Cleveland Prosecutor’s Office, DV advocates from the Witness/Victim Service Center, two attorneys from the Office of the Public Defender, and the 3 current judges on the Dedicated Domestic Violence Docket (see BLINDED REFERENCE for details). Although space restrictions prevent us from detailing all of the results of the qualitative assessment, some of the findings are referenced in our discussion of the quantitative findings presented here.
Results
Victim, Offender, and Case Characteristics
We identified and collected data on what we believe were all domestic violence cases within the City of Cleveland that met the DV Project definition of domestic violence (the victim was an adult female and the victim and defendant were either married, cohabitating, or had a child together) and occurred between July and December 2008. Although the number of cases on which we originally collected data was 1423, the final number of cases included in the analysis was 1388 after removing cases that were duplicates, handled by other agencies, or contained insufficient information. The largest percentage of cases came from the fourth (28.2%), fifth (25.4%) and second (23.2%) districts (that is, two out of three of the DV Project districts). Table 1 provides a breakdown of suspect and victim characteristics both in the sample as a whole as well as separately for cases handled by the DV Project and the first and fifth districts.
The majority of suspects for the entire sample of DV cases were Black males (77.9%) between the ages of 20 and 39 years (71.3%). The average age for all suspects was 32.7 years and was not significantly different for DV Project cases (M = 32.58, SD = 9.9) compared to non-DV Project cases (M = 32.92, SD = 10.7), t(873) = 0.575, p = .56. There was a higher percent of White suspects in cases handled by the DV Project (18.1%) than in the first and fifth districts (11.9%) and also a higher percent of Hispanic suspects (7.6% vs. 3.7%). Differences in suspect race across the two groups were statistically significant, χ2 (2, N = 1377) = 18.96, p < .01. Approximately 37% of all suspects had prior charges for domestic violence. The percent of suspects with prior domestic violence charges was significantly higher among DV project cases (46.6%) than non-DV Project cases (19.7%), χ2 (1, N = 1385) = 95.68, p < .01. Cases handled by the DV Project also had significantly higher percentages of suspects with prior domestic violence convictions (26% vs. 11.8%), χ2 (1, N = 1385) = 37.63, p < .01; and prior convictions for other violent offenses (35.5% vs. 19.1%) χ2 (1, N = 1385) = 39.93, p < .01. These patterns may be suggestive of greater charging and convictions in prior DV cases due to the existence of specialized units in those districts. About 45% of all suspects had prior drug convictions and this was not significantly different for DV Project and non-DV Project cases, χ2 (1, N = 1384) = 0.72, p = .40.
The majority of victims for the entire sample of DV cases were also Black (71.5%) and between the ages of 20 and 39 years (73.2%). For all victims, the average age was 30.74 years and was not significantly different for DV Project cases (M = 30.45, SD = 9.8) compared to non-DV Project cases (M = 31.31, SD = 10.6), t (885) = 1.46, p = .14. There was a higher percent of White victims in cases handled by the DV Project (24.4%) than in the first and fifth districts (18.0%) and also a higher percent of Hispanic victims (7.8% vs. 3.2%). Differences in victim race across the two groups were statistically significant, χ2 (2, N = 1380) = 21.03, p < .01. A breakdown of the suspect’s relationship to the victim reveals suspects were most often either a co-resident (33.5%) or boyfriend (27.6%) of the victim. It is important to note that determinations of the relationship status relied on whatever information was provided in the narrative of the police report and/or victim and witness statements. More systematic data collection on relationship type is among the recommendations resulting from this study. Very few victims had a current temporary protection order against the suspect (3.4%), although a significantly higher percent of cases handled by the DV Project (4.9%) involved existing temporary protection orders than cases handled by the first and fifth districts (0.6%), χ2 (1, N = 1364) = 16.49, p < .01.
Table 2 provides a breakdown of the case characteristics for the sample as a whole and also separately for DV Project and non-DV Project cases. The majority of incidents occurred in a joint residence of the suspect and victim. Incidents most commonly involved the suspect physically assaulting the victim (81%). Strangulation was also not uncommon (216 cases, or 15.6%). DV Project cases were significantly more likely to involve a knife (2.8% vs. 0.9%), χ2 (1, N = 1388) = 5.77, p = .016, assault (82.6% vs. 78.1%), χ2 (1, N = 1388) = 4.07, p = .044), or other type of weapon (7.7% vs. 4.0%), χ2 (1, N = 1388) = 6.99, p < .01, while non-DV Project cases were significantly more likely to involve strangulation (19.8% vs. 13.4%), χ2 (1, N = 1388) = 9.66, p < .01, blunt instrument (4.7% vs. 2.5%), χ2 (1, N = 1388) = 4.69, p = .03, or threats (16.4% vs. 9.5%), χ2 (1, N = 1388) = 14.23, p < .01. Almost half of victims received visible injuries (45%), although the percent was significantly higher for cases handled by the DV Project (50.7%) than non-DV Project cases (36%), χ2 (1, N = 1367) = 26.44, p < .01. It is possible this difference reflects a variation in reporting between the two groups. In particular, first responders in DV project districts may have a greater awareness of the importance of systematically and thoroughly documenting injuries in police reports given their relationship with DV unit detectives.
Case Follow-up for Domestic Violence Cases in the First and Fifth Districts
Of the 481 non-DV Project cases where the original incident occurred in the first and fifth districts, the overwhelming majority (75.3%) resulted in nothing beyond the initial incident report filed. In other words, the victim filed an initial report but did not follow up on the referral they were given to discuss the case with a City Prosecutor. No further action was taken on these 362 cases. In only 80 of the 481 cases (16.6%) did the victim pursue the case by following up on it at the Cleveland Prosecutor’s Office. The detective bureaus in the first and fifth districts (who handle felony domestic violence cases) handled 28 cases. Finally, the DV Project unit handled 11 of the cases, even though they did not occur in one of the DV Project’s district. Most often this occurred when the DV unit had previously handled a case involving the same suspect. These cases were subsequently included in the DV Project group as they were handled by DV detectives, prosecutors and judges.
Impact of DV Project on Prosecution Outcomes
There were 918 DV Project cases and 108 non-DV project cases reviewed by a prosecutor. Approximately half of these cases resulted in charges issued by the prosecutor, thus the initial attrition rate for DV Project cases was about 51% (Table 2). There were no significant differences in the percent of cases resulting in charges between the two groups (48.5% vs. 58.3%), χ2 (1, N = 1026) = 3.78, p =. 053. Slightly over three-quarters (76.3%) of the cases handled by the DV Project were heard by a Cleveland Municipal Court judge. A higher percent of cases from the first and fifth districts went through the Cuyahoga County Court of Common Pleas compared to cases handled by the DV Unit (36.5% vs. 23.7%), χ2 (1, N = 502) = 4.79, p = .03.
Table 3 displays the results of a logistic regression model analyzing factors predicting whether charges were issued by the Cleveland Prosecutor’s Office in domestic violence cases. With the exception of prior convictions for non-DV violent offenses and suspect and victim race, all of the predictors included in the model had a significant impact on the likelihood charges were issued in a domestic violence case. The strongest predictors were the victim having visible injuries, OR = 2.596, 95% CI [1.982, 3.401], and the existence of a temporary protection order, OR = 2.649, 95% CI [1.367, 5.132]. The lower likelihood of charges in cases handled by the DV Project remained significant with the addition of these predictors, OR = 0.499, 95% CI [0.321, 0.775]. Importantly, the odds of charges being issued were nearly twice as high in cases involving prior domestic violence charges against the defendant, OR = 1.868, 95% CI [1.414, 2.467]. Thus, even if only half of the cases result in charges, the high volume of cases being handled by the Cleveland Police Department’s DV Unit does and will continue to be a major factor in getting domestic violence cases into the court system.
Impact of the Dedicated Domestic Violence Docket on Court Outcomes
Table 4 displays the court outcomes for all of the domestic violence cases (defined according to the Violence Against Women Act criteria) handled by the Cleveland Municipal Court, including a comparison of cases handled by the DV Project (that were assigned to the Dedicated Domestic Violence Docket) and those handled by the first and fifth districts (that were assigned to one of the remaining judges on the court). For both groups of cases, the majority resulted in a plea of guilty or no contest by the defendant (64.6%). There was no significant difference between the two groups in the percent of cases where the defendant was found or plead guilty or no contest, χ2 (1, N = 350) = 1.57, p = .21.
The next most common outcome was case dismissal (22%). A significantly greater percent of cases handled by the DV Project were dismissed than those handled by the first and fifth districts (23.6% vs. 8.8%) χ2 (1, N = 348) = 3.87, p < .05. To assess whether the difference in case dismissals was a function of the DV Project or a reflection of the differences in the make-up of the samples, we ran a binary logistic regression model predicting the likelihood of case dismissal that included a measure of whether the case was from the DV Project or control group, along with whether the victim and defendant were living together, whether the victim had an existing temporary protection order against the suspect, whether the victim received a visible injury, whether the defendant had prior domestic violence changes, whether the defendant had a prior conviction for other violent offenses, whether the suspect was black, and whether the victim was black. The difference in case dismissals between DV Project and non-DV Project cases was no longer significant after controlling for these other variables (Table 5).
Only 12 cases resulted in a trial heard on the Dedicated Domestic Violence Docket, eight of which produced a guilty finding. None of the cases originating in the first and fifth districts resulted in the defendant being referred to a diversion program, while 11 defendants whose cases were handled by the DV Project received a referral. This may reflect a lack of familiarity with the Deferred Judgment Program among judges who are not on the Dedicated Domestic Violence Docket.
Approximately half of all of the cases resulted in a domestic violence conviction (48.5%). More than one in five cases resulted in an assault conviction (21.7%) and more than one in three cases produced a conviction for another type of offense (e.g. menacing). For defendants who were sentenced to serve time, the average sentence was 4.8 months. For the most common disposition, probation, the average sentence was 15.7 months. Few defendants received a fine (31 in total), but of those that did, the average fine cost was $300.32. None of the differences for convicted offenses or sentence length was significant for cases handled by the DV Project/ Dedicated Domestic Violence Docket compared to those in the non-DV Project districts.Footnote 3
With respect to defendants sentenced to treatment programs, a number of significant differences were found between DV Project and first/fifth district cases. Defendants whose cases were heard on the Dedicated Domestic Violence Docket were significantly more likely to be sentenced to the Domestic Intervention and Education Training program (DIET), χ2 (1, N = 244) = 3.88, p < .05, as well as other types of treatment programs (e.g. Get on Track, parenting classes etc.), χ2 (1, N = 215) = 5.57, p < .05, than those whose cases were not on the Dedicated Domestic Violence Docket. Furthermore, cases heard on the Dedicated Domestic Violence Docket were significantly less likely to result in a referral to anger management, χ2 (1, N = 212) = 7.86, p < .05, and substance abuse assessment and counseling, χ2 (1, N = 212) = 7.45, p < .05. It appears non-Dedicated DV judges may be sentencing domestic violence offenders to anger management programs instead of batterer intervention. These differences in dedicated and non-dedicated docket cases in the likelihood of referrals to DIET and anger management remained significant when we conducted binary logistic regression models that controlled for other case characteristics.
We analyzed the court outcome data a second time with the goal of comparing all of the judges who had been assigned to the Dedicated Domestic Violence Docket with those who had never been assigned to it as of the time the data were collected. We anticipated that the training and experience of handling domestic violence cases on the dedicated docket would continue to have an impact even after judges had rotated off the docket. The same significant differences emerged between the dedicated docket and non-dedicated docket judges as we found previously when comparing the DV Project cases with those cases from the first and fifth districts. The similarity in results likely stems from the fact that so few cases that originate in the first and fifth districts ever make it as far as the Cleveland Municipal Court and thus even judges who were previously assigned the Dedicated Domestic Violence Docket but have since rotated off hear very few domestic violence cases.
When comparing the percent of defendants sentenced to prison among cases handled by Dedicated Domestic Violence Docket judges versus other judges, we found that among the dedicated docket judges 22.2% of defendants were sentenced to prison compared to 44.4% of defendants whose cases were handled by other judges. This difference, which is statistically significant, χ2 (1, N = 228) = 5.73, p < .05, may indicate that Dedicated Domestic Violence Docket judges were attempting to use incarceration as a last resort in DV cases and instead were sentencing defendants to probation where they can be referred to a range of services and programs. In contrast to prison sentences, for cases sentenced to probation we found no significant difference between the two groups in terms of the length of probation sentences given, t(210) = 0.742, p = .46.
Impact of the DV Project and Dedicated Domestic Violence Docket on Offender Recidivism
In order to examine the impact of the DV Project and Dedicated Domestic Violence Docket on the commission of additional domestic violence offenses by defendants, we collected information on the number of domestic violence incidents reported to the Cleveland Police Department for each defendant in our sample for a period of two to two and a half years after the initial domestic violence incident (through December 31, 2010). Of the 1279 defendants, 398 (31.1%) involved a defendant on whom subsequent domestic violence incident reports were filed with the Cleveland Police Department. Because we were not able to obtain police reports for surrounding suburbs, and we only had data on reports titled domestic violence (and not, for example, child endangerment or menacing), we view this as a conservative estimate of the degree of recidivism among domestic violence perpetrators.
Of those defendants for whom additional domestic violence incident reports were identified, 63% had only one additional report. In 19% of cases there were two additional DV incidents reported. Although 90% of cases involved three or fewer subsequent DV incidents, we identified a handful of extreme cases. For example, there were two defendants for whom eight additional domestic violence reports existed, and two for whom there were seven subsequent reports.
We compared DV project cases with non-DV project cases in terms of the percent of cases involving the commission of additional domestic violence offenses by defendants. No significant differences were found, χ2 (1, N = 1279) = 0.43, p = 0.51. We also compared defendant recidivism for DV project/Dedicated DV Docket cases and non-DV project cases for just those cases that were heard by the Cleveland Municipal Court. Again, the differences were not significant, χ2 (1, N = 350) = 0.30, p = .58.
Discussion and Policy Implications
Our evaluation of the City of Cleveland’s Domestic Violence Project and the Cleveland Municipal Court’s Dedicated Domestic Violence Docket has produced an array of conclusions and policy implications. The most important of these is the overall positive impact of the police DV unit and the DV advocates from the Witness/Victim Service Center on the handling of domestic violence cases in the City of Cleveland. Our findings demonstrate that the traditional way law enforcement handles misdemeanor DV cases, where the onus is on the victim to meet with a city Prosecutor, is ineffective to the extent that only a very small proportion of victims actually follow-up on their cases with a Prosecutor. Thus, in districts that lack the DV Project, the overwhelming majority of domestic violence cases result in nothing further than an initial police report, emphasizing the significant impacts police DV units have on moving misdemeanor domestic violence cases forward in the city. A substantially greater number of investigations of DV cases occur in the districts with the DV units, demonstrating that the DV Project has been successful in meeting that particular objective.
The low numbers of victims with temporary protection orders is cause for concern. It is possible that we are underestimating this percent, as we had to rely on this information being recorded in the police report. However, it is also possible that victims in the first and fifth district are not meeting with a city prosecutor to pursue obtaining a protection order because they do not wish to see the domestic violence case go forward. More generally it is possible that victims do not believe obtaining a temporary protection order will be effective or they anticipate resuming their relationship with the suspect at some point.
Through the work of those in the DV Project, the DV prosecutors in the Cleveland Prosecutor’s Office were able to issue charges in nearly half of all of the domestic violence cases that come through the Project. Comparing the rate of charges issued in Cleveland with elsewhere, recent studies show that prosecutors dismiss about half of intimate partner violence cases with no charges issued (Belknap et al., 2010). Thus, charging patterns in Cleveland do not seem to differ from those found elsewhere. However, given that one of the goals of the DV Project is to increase offender accountability, the lower likelihood of cases resulting in charges for the DV Project suggests that this goal, and the more specific objective to increase the number of charges in DV cases, have not yet been met. Cases being handled by the DV Unit were more likely to involve a suspect who had prior DV charges and convictions and prior convictions for other violent offenses. This pattern can be viewed as evidence of the positive impact the DV Project is having where DV cases within the Project are more likely to result in domestic violence charges and convictions that in turn can result in escalating charges in future incidents.
Domestic violence cases originating in the non-DV Project districts that make it to court had a greater likelihood of charges that are felony-level. This may occur for several reasons. First, approximately one-quarter of the cases from these districts that resulted in some kind of follow-up were handled by the districts’ detective bureaus, which only handle felony cases, which are naturally likely to produce felony charges. Second, victim cooperation may be greater in cases from the first and fifth districts that make it as far as the court system as two-thirds of the cases from the control group that resulted in follow-up were pursued by the victims themselves, who took the time and effort to go downtown and meet with a city prosecutor to discuss the case. This will increase the likelihood of felony charges since cooperation should facilitate putting together a stronger case including more serious charges, and make it more probable that a victim will pursue charges for repeated incidents, leading to escalated charges.
For cases handled by the Cleveland Municipal Court, the majority of cases handled by the Dedicated Domestic Violence Docket resulted in findings or pleas of guilty or no contest. Although the rate of case dismissals was higher for DV Project than non-DV Project cases, this difference did not remain significant once we controlled for differences in the samples. Evaluations of other specialized court programs reveal relatively high rates of domestic violence cases being dismissed or nollied due to a lack of victim cooperation (e.g., Buzawa, Hotaling, Klein, & Byrne, 1999). That said, research on other dedicated domestic violence dockets suggests that dedicated dockets can reduce dismissal rates in domestic violence cases (Karan, Keilitz, & Denaro, 1999). Our finding that DV Project cases do not have lower dismissal rates than non-DV Project cases suggests that the goals of increasing offender accountability through a greater reliance on evidence-based prosecution and the objective to increase successful prosecutions are not currently being met. There are several plausible explanations for this finding. Focus groups conducted with DV Prosecutors and advocates about the DV Project and Dedicated Domestic Violence Docket revealed several obstacles to the successful prosecution of cases handled by the DV Project. For example, both the DV Prosecutors and advocates reported feeling overwhelmed by their caseloads, which both groups viewed as extremely high.
There was also widespread consensus among the DV advocates that the current structure of the Dedicated Domestic Violence Docket results in victims having to spend a lot of time sitting around in court waiting for their cases to be called. This poses a hardship on victims, who are told to come to court at 9:00 a.m. but their cases are often not heard until hours later. The advocates report that the inconvenience this causes victims, who are frequently juggling childcare and work issues with attending court, leads them to become disheartened by the process, reduces victim cooperation and increases the likelihood the victim will recant in order to hasten an end to the case. This problem is compounded by the fact that many victims receive subpoenas the day before they are supposed to appear in court and upcoming court dates are not appearing in the Court website in a timely fashion. Domestic violence victims face a number of hurdles in showing up to court for their cases. Among the obstacles they face are securing time away from employment and finding alternatives for childcare. Their ability to overcome these issues and be present in court would be greatly facilitated by as much advanced notification as possible that their presence is going to be required in court. Given the reluctance of many victims to participate in the court process, making this aspect less burdensome to them by providing subpoenas in a timely fashion is an important priority.
The DV advocates also identified a number of obstacles to following up with victims based on initial police reports, including incomplete reports, victims’ addresses sometimes missing from the report or the lack of a phone number for the victim. They stressed the importance of including a contact number in the report, even if the number is for a neighbor or relative rather than the victim. They also noted that first responders in the second, third, and fourth districts sometimes incorrectly refer victims to the City Prosecutor’s Office when they should be referring them to the DV unit or the Witness/Victim Service Center. As a result, victims are wasting time sitting around in the Cleveland Prosecutor’s Office only to find out that they are in the wrong place. These concerns highlight the need for better training for first line officers in writing police reports.
Judges on the Dedicated Domestic Violence Docket appear to rely less on prison in dealing with convicted defendants and more likely to require defendants to participate in programming such as batterer intervention as part of their probation sentence. These practices may require re-examination in light of research demonstrating lower re-arrest rates among domestic violence defendants given more intrusive sentences (jail, work release etc.) and the weak and inconsistent findings with respect to the effectiveness of batterer intervention programs (Auchter & Backes, 2013; Babcok, Green, & Robie, 2004; Daly & Pelowski, 2000; Feder & Forde, 2000; Feder & Wilson, 2005; Venture & Davis, 2005). Judges should be educated with respect to these research findings. Furthermore, comparing the recidivism rates of defendants in our sample sentenced to the local batterer intervention program (DIET) with those who were not, we found no statistically significant difference in whether defendants committed subsequent domestic violence offenses.Footnote 4
The recidivism rate among domestic violence defendants is not significantly different for DV Project cases compared to those from the control group, which does raise questions with respect to whether the DV Project is meeting its goal of enhanced offender accountability. High rates of recidivism among domestic violence offenders have been reported in many evaluations of court-based strategies for improving the handling of domestic violence cases (Ursel & Hagyard, 2008). Thus, recidivism among domestic violence offenders has long been a problem for the courts. Cleveland seems to be no exception. Closer scrutiny and evaluation of the services provided to defendants convicted of domestic violence may be needed to identify areas for improvement beyond what is occurring up to and including the court proceedings.
The results of this evaluation study point to a number of additional policy implications. Perhaps most importantly, the value of implementing domestic violence units in police departments is strongly supported by the results. This is an effective model for getting initial police reports on domestic violence offenses to move beyond this first step. However, for the positive impact of specialized domestic violence units to occur throughout all of the stages of the criminal justice process, each specialized unit must be appropriately staffed. A significant challenge for the current system in Cleveland is the overwhelming caseloads shouldered by only a handful of DV prosecutors in the Cleveland Prosecutor’s Office. For the 6-month period examined here, the Cleveland Prosecutor’s Office reviewed around 1000 DV cases. About half of these cases resulted in charges and thus additional work for the DV prosecutors. Given that rates of domestic violence both in Cleveland and elsewhere have increased during the past several years, the number of cases currently being handled by DV prosecutors has continued to grow. The difficulties this creates for building strong, evidence-based cases that do not rely on victims showing up to court are further compounded by the understaffing of the DV detective unit and the need for more DV advocates. Thus, for jurisdictions considering adopting this type of model for handling domestic violence offenses, sufficient staffing will be critical.
Our findings emphasize the importance of providing adequate training to law enforcement officers regarding the documentation of injuries in domestic violence cases. Even for victims who do not follow up with prosecutors, well-written police reports provide the bases for stronger cases should subsequent incidents occur. Furthermore, extensive documentation of injuries can increase the likelihood of evidence-based prosecution.
The study results highlight the need for victim education on the necessary steps for obtaining a protection order. Having a domestic violence unit to handle DV cases agency-wide may assist in addressing this gap. If all DV victims are assigned an advocate from the time the police report is initially filed, the advocate could explain the steps required to obtain a protection order as well as the value of doing so.
An important consideration in structuring a dedicated domestic violence docket is to have DV cases scheduled to be heard at specific times of the day (e.g. 11 a.m. until 1 p.m.) on specific days (2 or 3 days a week). This could eliminate the time wasted by victims, prosecutors, detectives, and advocates wait hours for the case to be called. Ultimately this should also improve victim cooperation by creating some predictability for the victims regarding when they will need to be in court. If possible, the scheduling of DV cases should be staggered across those judges with the DDVD so that DV detectives and advocates are not scheduled to be in multiple courtrooms at the same time.
Finally, the creation of a specialized DV unit in the Office of the Public Defender will provide some needed balance to a system that has specialized DV prosecutors. It should also improve the ability of public defenders to adequately represent clients involved in DV cases.
While the current study uses a stronger methodological design than many evaluations of specialized policing units and courts, it is not without its limitations. The most significant of these was the reliance on the accuracy of the information in the police reports. In some cases, it was difficult to determine the exact nature of the relationship between the victim and the offender. Information on protection orders may not have been consistently reported in the police reports. And we had to rely on a conservative measure of recidivism that excluded police reports for surrounding suburbs and reports titled something other than domestic violence (e.g., child endangerment or menacing). Finally, our study was limited to a single jurisdiction. Further research is needed using similar quasi-experimental designs to assess the generalizability of the findings reported here.
Notes
Estimates at the police district level were derived from the U.S. Census Bureau’s 2010 Census and the American Community Survey, 2007–2011, and were produced by the Northern Ohio Data & Information Service (NODIS), Maxine Goodman Levin College of Urban Affairs, Cleveland State University.
Although we would have liked to collect information to measure economic status, such as income, this was not available in the files.
Domestic violence conviction χ2 (1, N = 235) = 0.26, p = .87; assault conviction χ2 (1, N = 235) = 3.377, p = .07; conviction for other offenses χ2 (1, N = 235) = 3.170, p = .08; prison sentence length t(72) = −0.507, p = .61; probation sentence length t(210) = 0.742, p = .46.
Unfortunately we did not have information on whether defendants sentenced to attend batterer’s intervention actually successfully completed the program.
References
Auchter, B., & Backes, B. L. (2013). NIJ’s program of domestic violence research: Collaborative effects to build knowledge guided by safety for victims and accountability for perpetrators. Violence Against Women, 19, 713–736.
Babcok, J. C., Green, C. E., & Robie, C. (2004). Does batterers’ treatment work? A meta-analytic review of domestic violence treatment. Clinical Psychology Review, 23, 1023–1053.
Belknap, J., Hartman, J. L., & Lippen, V. L. (2010). Misdemeanor domestic violence cases in the courts: A detailed description of the cases. In V. Garcia & J. E. Clifford (Eds.), Female victims of crime: Reality reconsidered (pp. 259–278). Upper Saddle River, NJ: Prentice Hall.
Bennett, L., Goodman, L., & Dutton, M. A. (1999). Systemic obstacles to the criminal prosecution of a battering partner: A victim perspective. Journal of Interpersonal Violence, 14, 761–772.
Black, M. C., Basile, K. C., Breiding, M. J., Smith, S. G., Walters, M. L., Merrick, M. T., Chen, J., & Stevens, M. R. (2011). The national intimate partner and sexual violence survey (NISVS): 2010 summary report. Atlanta, GA: National Center for Injury Prevention and Control, Centers for Disease Control and Prevention.
Browne, A. (1997). Violence in marriage: Until death do us part? In A. P. Cardarelli (Ed.), Violence between intimate partners: Patterns, causes, and effects (pp. 48–69). Needham Heights, MA: Allyn and Bacon.
Buzawa, E. S., & Buzawa, A. D. (2008). Courting domestic violence victims: A tale of two cities. Criminology & Public Policy, 7, 671–685.
Buzawa, E. S., Buzawa, C. G., & Stark, E. (2012). Responding to domestic violence: The integration of criminal justice and human services. Los Angeles: Sage Publications.
Buzawa, E. S., Hotaling, G. T., Klein, A., & Byrne, J. (1999). Response to domestic violence in a pro-active court setting: Final report. Lowell: University of Massachusetts.
Catalano, S. (2007). Intimate partner violence in the United States. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.
Catalano, S. (2012). Intimate partner violence, 1993–2010. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.
Daly, J., & Pelowski, S. (2000). Predictors of dropout among men who batter: A review of studies with implications of research and practice. Violence and Victims, 15, 137–160.
Davis, R. C., O’Sullivan, C. S., Farole Jr., D. J., & Remple, R. (2008). A comparison of two prosecution policies in cases of intimate partner violence: Mandatory case filing versus following the victim’s lead. Criminology & Public Policy, 7, 633–662.
Davis, R. C., Smith, B. E., & Nickles, L. (1997). Prosecuting domestic violence cases with reluctant victims: Assessing two novel approaches in Milwaukee. Report to the National Institute of Justice.
Dawson, M., & Dinovitzer, R. (2001). Victim cooperation and the prosecution of domestic violence in a specialized court. Justice Quarterly, 18, 593–622.
Dugan, L., Nagin, D. S., & Rosenfeld, R. (1999). Explaining the decline in intimate partner homicide: The effects of changing domesticity, women’s status, and domestic violence resources. Homicide Studies, 3, 187–214.
Dugan, L., Nagin, D. S., & Rosenfeld, R. (2003). Exposure reduction or retaliation? The effects of domestic violence resources on intimate-partner homicide. Law & Society Review, 37, 169–198.
Erez, E., & Belknap, J. (1998). In their own words: Battered women’s assessments of the criminal processing system’s responses. Violence and Victims, 13, 251–268.
Feder, L., & Forde, D. (2000). A test of the efficacy of court-mandated counseling for domestic violence offenders: The Broward experiment. Final report for the National Institute of Justice, grant number 96-WT-NX-0008. Washington, DC: U.S. Department of Justice, National Institute of Justice. NCJ 184752.
Feder, L., & Wilson, D. B. (2005). A meta-analytic review of court-mandated batterer intervention programs: Can courts affect abusers’ behavior? Journal of Experimental Criminology, 1, 239–262.
Friday, P. C., Lord, V. B., Exum, M. L., & Hartman, J. (2006). Evaluating the impact of a specialized domestic violence police unit, final report. Washington, DC: National Institute of Justice.
Garcia, V., & McManimon, P. (2011). Gendered justice: Intimate partner violence and the criminal justice system. Lanham, MA: Rowman & Littlefield Publishers.
Iyengar, R. (2009). Does the certainty of arrest reduce domestic violence? Evidence from mandatory and recommended arrest laws. Journal of Public Economics, 93, 85–98.
Johnson, H., & Dawson, M. (2011). Violence against women in Canada: Research and policy perspectives. Don Mills, Ontario: Oxford University Press.
Jones, D. A., & Belknap, J. (1999). Police responses to battering in a progressive pro-arrest jurisdiction. Justice Quarterly, 16, 249–273.
Karan, A., Keilitz, S., & Denaro, S. (1999). Domestic violence courts: What are they and how should we manage them? Juvenile and Family Court Journal, 50, 75–86.
Labriola, M., Bradley, S., O’Sullivan, C. S., Rempel, M., & Moore, S. (2009). A national portrait of domestic violence courts. Washington, DC: National Institute of Justice.
Lauritsen, J. L., Owens, J. G., Planty, M., Rand, M. R., & Truman, J. L. (2012). New methods for counting series victimizations. Washington DC: U.S. Department of Justice, Bureau of Justice Statistics.
Newmark, L., Rempel, M., Diffily, K., & Kane, K. M. (2001). Specialized felony domestic violence courts: Lessons on implementation and impacts from the kings county experience. Washington DC: The Urban Institute.
Orchowsky, S. J. (1999). Evaluation of a coordinated community response to domestic violence intervention: The Alexandria domestic violence intervention project. Final report. Washington DC: National Institute of Justice.
Peterson, R. R. (2008). Reducing intimate partner violence: Moving beyond criminal justice interventions. Criminology & Public Policy, 7, 537–545.
Robinson, A. L., & Maxwell, C. D. (2008). The challenge to respond effectively to violence against women in a global context. International Journal of Comparative and Applied Criminal Justice, 32, 133–147.
Shelton, D. E. (2007). The current state of domestic violence courts in the United States, 2007. National Center for State Courts White Paper.
Sigler, R. T., Crowley, J. M., & Johnson, I. (1990). Judicial and prosecutorial endorsement of innovative techniques in the trial of domestic abuse cases. Journal of Criminal Justice, 18, 443–453.
Singer, S. I. (1988). The fear of reprisal and the failure of victims to report a personal crime. Journal of Quantitative Criminology, 4, 289–302.
Spohn, C. (2008). Editorial introduction to coordinated community response to intimate partner violence. Criminology and Public Policy, 7, 489–493.
Stark, E. (2007). Coercive control: How men entrap women in personal life. New York: Oxford University Press.
Townsend, M., Hunt, D., Kuck, S., & Baxter, C. (2005). Law enforcement response to domestic violence calls for service. Final report for National Institute of Justice, grant number 99-C-008. Washington, DC: U.S. Department of Justice, National Institute of Justice NCJ 215915.
Truman, J. L., & Morgan, R. E. (2016). Criminal Victimization, 2015. Washington, DC: Bureau of Justice Statistics, U.S. Department of Justice.
Ursel, J., & Hagyard, C. (2008). The Winnipeg family violence court. In J. Ursel, L. M. Tutty, & J. Lemaistre (Eds.), What’s law got to do with it? The law, specialized courts, and domestic violence in Canada (pp. 95–119). Toronto: Cormorant Books.
Venture, L., & Davis, G. (2005). Domestic violence: Court case conviction and recidivism. Violence Against Women, 11, 255–277.
Xie, M., Lauritsen, J. L., & Heimer, K. (2012). Intimate partner violence in U.S. metropolitan areas: The contextual influences of police and social services. Criminology, 50, 961–992.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
About this article
Cite this article
Regoeczi, W.C., Hubbard, D.J. The Impact of Specialized Domestic Violence Units on Case Processing. Am J Crim Just 43, 570–590 (2018). https://doi.org/10.1007/s12103-018-9435-9
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s12103-018-9435-9