Factors Influencing Law Enforcement Responses to Child to Parent Violence

Abstract

A domestic violence incident perpetrated by a child toward his or her parent presents a challenging dynamic for law enforcement officers responding to these calls for service. To date, law enforcement responses to child to parent violence (CPV) have only been studied dichotomously (i.e., decision to arrest), and as a result, the associated complexities are not well understood. Here, we add to the understanding of individual, situational, and contextual factors that influence law enforcement response to CPV by examining 1,113 calls for service in a Midwestern state. In assessing the relative influence of these factors on responses using a multinomial logistic regression with cluster robust standard errors, we find evidence that the gendered nature of CPV victim-offender dyads and the presence of victim injury influences police decision to arrest in lieu of an informal, de-escalation only response. We also find officers are less likely to refer youth to social welfare agencies or arrest a youth perpetrator when CPV occurs in neighborhoods with a high level of family disruption. In these instances, the officer works to deescalate the situation, but is significantly less likely to take further action or formally refer the family to social resources. Researchers must work to better understand CPV prevention and improve system responses and resources for affected families.

Keywords

youth violence legal intervention domestic violence predicting domestic violence

Domestic violence (DV) incidents perpetrated by a child against his or her parent present a challenging situation for law enforcement officers. Child to parent violence (CPV) situations are complex due to the victim-offender relationship, legal status of the offender, and many other factors. When responding to a call for service in a CPV situation, officers initially focus on deescalating the violent incident. Once the officer has successfully deescalated the CPV incident, some studies find parent victims are reluctant to engage further with law enforcement officials. Researchers suggest this reluctance may partially be due to embarrassment and fear of being judged (Agnew & Huguley, 1989), worry about being blamed by others for their child’s behavior (Moulds, Day, Mildred, Miller, & Casey, 2016), or feelings of self-blame for not being able to control their child (Cohen-Filipic & Bentley, 2015; Williams, Tuffin, & Niland, 2017). Such victim reluctance to engage presents a challenge to officers responsible for determining the most appropriate law enforcement response. In some jurisdictions, a lack of law enforcement policy guiding response to CPV can result in over reliance on discretionary decision making by responding officers (Miles & Condry, 2016). Given these complexities, researchers must determine how situational, organizational, and other contextual factors surrounding the CPV incident contribute to the officer’s decision (Buzawa & Hotaling, 2006).

Prior research on CPV is limited and focuses almost exclusively on developing victim and offender typologies. When assessing law enforcement responses, studies are limited to examining a dichotomous outcome of arrest decision. Given this nature of the current CPV literature, we lack a nuanced understanding of responding to CPV in our communities to systematically balance safety and welfare needs (Moulds et al., 2016). The logical starting point is to develop a better understanding of officer decision making in CPV incidents. Specifically, when do officers invoke a nonarrest response option such as a referral to a social welfare agency for family assistance in comparison to making an arrest? The purpose of the study presented here is to better understand factors that account for the variation in law enforcement responses to children who reportedly engage in parental violence. We begin by contextualizing this study within the CPV literature. Specifically, we emphasize similarities and distinctions between CPV and DV incidents followed by discussion of existing empirical literature on police response to CPV.

CPV

In response to legal action, public pressure, and empirical research that called into question prior police inaction in adult DV incidents, law enforcement agencies enacted numerous mandatory or preferred arrest policies throughout the United States (Police Executive Research Forum [PERF], 2015). In addition to policies advocating for arrest, other common practices adopted by law enforcement agencies include warrantless arrest, specialized law enforcement units and training, on-scene assessment of lethality, and provision of assistance to victims with protection orders (PERF, 2015). Fundamental changes to police response to DV in general has resulted in an overall increase in arrest rates for all DV perpetrators including CPV perpetrators (Buzawa & Hotaling, 2006).

Domestic incidents not involving intimate partners, including CPV, have not received the same level of attention as intimate partner violence. In reference to definitions for domestic or family violence, there is little mention of emotional or physical abuse committed by children against their parents. For instance, the National Criminal Justice Reference Services (NCJRS) describes family violence as “physical, sexual, financial abuse, child abuse and neglect, and elder abuse” (NCJRS, 2017), and while the NCJRS website includes resources available to better understand family violence within the context of intimate partner violence, child abuse, children exposed to violence, and elder abuse, resources for parents experiencing CPV are absent.

Definition and Prevalence of CPV

In the 1970s, researchers initially recognized “battered parent syndrome” as a discrete form of family violence (see Harbin & Madden, 1979). Contemporary literature more clearly specifies the child as the instigator of violence toward his or her parent using CPV terminology. Still, no universal definition of this type of violence exists. CPV terminology broadly encompasses acts of violence committed by a minor child against a parent. As one example, Cottrell and Finlayson (1996) specify CPV as “any act of a child that is intended to cause physical, psychological, or financial damage to gain power and control over a parent” (p. 3).

The lack of a consistent CPV definition, variation in the age range of study samples and differing methods employed in data collection (surveys, interviews, file reviews, case studies, and official data) in CPV studies have created difficulties in determining the exact prevalence of CPV in our communities (Moulds et al., 2016). Recent studies estimate CPV incidents in the United States comprise 7% to 18% of DV incidents in dual parent households and 29% of DV incidents in single parent households (Erez & Tontodonato, 1989; Kennair & Mellor, 2007; Walsh & Krienert, 2009). This already high prevalence of CPV in families is likely underestimated. Parental reluctance to report CPV as described earlier likely results in a prevalence estimate much lower than actually exists (Correll, Walker, & Edwards, 2017).

The vast majority of the CPV literature focuses on profiling individuals who CPV affects. Researchers who engage in efforts to better document this type of violence have examined youth and family characteristics, risk factors, and motivations for engaging in such violence (Agnew & Huguley, 1989; Calvete, Orue, & Gamez-Guadix, 2013; Calvete, Orue, Gamez-Guadix, & Bushman, 2015; Condry & Miles, 2014b; Contreras & Cano, 2014; Ibabe & Bentler, 2016; Lyons, Bell, Frechette, & Romano, 2015; Routt & Anderson, 2011; Walsh & Krienert, 2009). Generally, findings indicate CPV incidents share many characteristics with adult DV (Holt, 2016). For instance, CPV is a gendered crime such that females are the most common victim and males are the most common offender (Condry & Miles, 2014a). Using National Incident-Based Reporting System (NIBRS) data, Strom, Warner, Tichavsky, and Zahn (2014) confirmed this similarity between DV and the CPV incidents. They noted males (i.e., sons) comprised 60% of offenders in CPV incidents, while females (i.e., mothers) comprised 74% of victims. In addition, Strom and colleagues found the vast majority of CPV incidents reported occurred inside the home (93%) and with a weapon (80%). About one third of victims in CPV incidents sustained an injury. Other similarities between DV and CPV include poly-victimization, short- and long-term physical, psychological, economic and social consequences, and victim blaming (Holt, 2016; Routt & Anderson, 2011).

For as many shared characteristics between CPV and DV, there are several distinguishing characteristics between these two types of violence, including the unique tactics used by perpetrators, the interaction of parenting–victim blaming (Cohen-Filipic & Bentley, 2015), the role of mental illness (Kennedy et al., 2010; Kethineni, 2004), and the potential for CPV to act as a pathway into adult DV for adolescent offenders (Holt, 2016; Routt & Anderson, 2011). Other risk factors that may uniquely increase the likelihood of CPV include a family history of child maltreatment, parental DV, delinquent peers, exposure to media violence, gender role socialization, and change in family structure (see Hong, Kral, Espelage, & Allen-Meares, 2012 for a review). Finally, Contreras and Cano (2015) among others find CPV perpetrators show a high level of psychopathology and social-cognitive difficulties (see also Agnew & Huguley, 1989). It is also important to recognize that unlike adult DV, in CPV the victim (i.e., parent) has a legal responsibility for the offender (i.e., child), which further complicates the law enforcement response to CPV.

Law Enforcement Perspectives on CPV

Unlike other subtypes of DV (e.g., intimate partner violence and child abuse), law enforcement response to CPV is understudied and as a result not well understood. The unique characteristics of CPV affect the manner in which law enforcement and social service agencies become aware of, and respond to this form of DV. The age of the offender (i.e., juvenile) and the relationship between the offender and victim (i.e., child-parent) call into question the manner in which the criminal justice system should respond. Uncertainty about whether CPV is a juvenile justice problem, a child protection issue, or a subtype of DV affects policy (Holt, 2016).

Local law enforcement perspectives tend to guide discretionary responses when clear policies are otherwise lacking, especially regarding appropriation of social resources across communities (Miller, 2015). Jurisdictions that view CPV through a DV lens wherein “any and all abusive responses to conflict, regardless of intent or impact, are defined as domestic violence” (Holt, 2016, p. 491) may be less likely to expend social resources on the matter. In these instances, arrest is the response of choice. This response to youth involved in CPV is similar to other violent incidents where the victim and offender have a narrow social distance (i.e., intimate partner violence; Rollwagen & Jacob, 2018). The intent of such arrest-based policies is to reduce crime-specific recidivism and hold offenders accountable for their actions. Limited attention is given to the impact of such policies on other less recognized forms of DV including CPV. Instead, arrest-based policies communicate to perpetrators and the community that this type of violence will not be tolerated, reduces officer discretion, and may result in more equitable law enforcement responses.

Officers in jurisdictions that view CPV as a juvenile justice or child protection issue tend to be less likely to arrest youth perpetrators and more likely to consider informal handling of the situation. From this perspective, arrest is viewed as criminalizing youth who are “acting out” and therefore not a feasible option. Here, CPV is thought to be a product of early experiences with hostile and aggressive interactions between children and parents (Kennedy et al., 2010). As these “troubled” kids may outgrow the problem, proponents argue it would be better to divert the youth or manage the situation informally.

Furthermore, parents who perceive the youth as “merely acting out” may be reluctant to initiate or follow through on formal processes (e.g., arrest and criminal justice-related processing) that will criminalize their child. Qualitative research in the United Kingdom specifically supported the finding that police and parents prefer a diversionary approach to arrest (Miles & Condry, 2016). This approach could involve a referral to a social welfare agency outside of the criminal justice system. An added concern about arrest is initiating youth involvement into the criminal justice system. Arguably, such involvement may result in delinquent persistence, thereby initiating a pathway to adult offending.

There is some indication that CPV has a higher probability of occurring in households where DV is already present (e.g., a parent is abusing child, parents engaged in violence) as well as parental drug use (Calvete, Orue, Gamez-Guadix, & Bushman, 2015; Pelletier & Coutu, 1992). Although police intervention and arrest may be necessary to break this intergenerational transmission of violence (Gebo, 2007), scholars argue criminalizing youth violence through arrest and criminal justice processing in these circumstances may not be appropriate in all cases, especially if youth are reacting to their own history of victimization.

Unfortunately, CPV studies in the United States tend to examine law enforcement responses using a dichotomous measure of arrest. This measurement approach inadequately captures the extent to which officers use other informal approaches (e.g., referral to social resources) as compared with a nonarrest response wherein only efforts to deescalate the incident occurs. Instead, researchers combine these possible approaches and compare them with a preferred arrest policy. Our understanding of factors that influence various response options beyond arrest/nonarrest are limited as a result.

Law Enforcement Response to CPV

Some studies suggest the majority of CPV incidents that come to the attention of the police result in arrest, whereas other studies find only half of such incidents result in arrest. Specifically, Miles and Condry (2016) find U.K. police arrested the child perpetrator in 94.6% of the incidents noted among the 100 U.K. police files reviewed. In a U.S.-based study, Strom and colleagues (2014) examine 54,197 domestic assaults with a juvenile perpetrator and parent victim reported in NIBRS between 2000 and 2004. They find in 55.5% of CPV cases the juvenile offender was arrested. The need to identify factors that explain this extant variation in decision to arrest is a critical knowledge gap in our understanding of police response to CPV.

The similarities between CPV and DV suggest researchers examining CPV can consider models developed in the DV literature to improve our understanding of variation in law enforcement responses. In addition to the impact of arrest policy (i.e., mandatory or preferred arrest policies), prior literature finds several factors influence law enforcement response to non-CPV specific DV including individual (e.g., victim/offender characteristics), situational (e.g., presence of injury), and contextual (e.g., social disorganization) factors. Yet, some distinct factors influencing police response may also exist given the unique aspects of the victim-offender relationship in CPV. Strom and colleagues (2014) find the odds of arrest in CPV incidents were significantly greater in jurisdictions with mandatory or preferred arrest policies, with female victims (i.e., assaults against mothers), and when the incident involved weapons, presence of victim injury, other offenses, and the perpetrator’s use of alcohol or drugs.

DV research finds that males are more likely than females to cause physical injury to their intimate partners (Felson & Cares, 2005) and when they do, the police are more likely to arrest them (Dichter & Rhodes, 2011; Hamilton & Worthen, 2011). In cases of CPV, however, prior studies indicate regardless of the sex of the perpetrator (i.e., the child), mothers are more likely to be the victim than fathers are (Erez & Tontodonato, 1989; Strom et al., 2014), so the impact of injury on arrest is unknown. Many other questions remained unanswered about the complex dynamics of CPV and law enforcement response. Studies find the perpetrator’s gender may influence how law enforcement interprets the incident (i.e., the seriousness of the event). Less is known about whether the gender pairing of the victim and offender influences police responses to CPV incidents. Does the gender of the perpetrator and the gender of the victim influence the police action taken? Does the same threshold of physical injury resulting arrest exist in cases of CPV as found in the DV literature?

Research examining arrest decisions in non-CPV specific juvenile delinquency may shed some light on the importance of other contextual factors important for CPV. Underscoring the importance of including contextual assessment in analysis of officer discretion, Schulenberg (2003) suggests “ecological contamination” exists in which an officer’s perception of a community influences discretion in their interactions with youth. Using data from Canadian municipalities in cases involving juveniles, Schulenberg (2003) finds social disorganization variables, but not officer workload, affects the use of formal police action. In examining CPV incidents, Strom et al. (2014) find a significant, positive effect of population size on the likelihood of juvenile arrest. Expanding the understanding of contextual factors in CPV following this early line of investigation is worthy of further consideration. Specifically, how might the level of family disruption in a community influence police interpretation and response to CPV?

Current Study

There is a paucity of research on factors that affect the law enforcement response (i.e., decision to arrest) regarding the child perpetrator in a CPV incident. The current study aims to determine the influence of individual, situational, and contextual factors on police response to incidents of violence perpetrated by a child toward his or her parent(s) relative to other police response options. Alternatives to arrest include handling the incident informally (no action beyond de-escalation of the event) or referring the child to a social welfare or other agency. This study advances the limited existing literature in three important ways. First, recognizing that parents often prefer diversionary responses, this study expands the outcome variable beyond the commonly used dichotomous approach to include the three possible law enforcement responses described below (i.e., de-escalation only, referral to social services, or arrest). Second, we expand upon typical individual factors considered to include gender dyads predicated on the relationship between the victim (i.e., parent) and offender (i.e., child). Prior literature has distinctly considered the gender of the victim and offender (Strom et al., 2014), but has not used a dyadic measurement approach. Finally, an additional contextual consideration of the level of family disruption within the jurisdiction that the CPV incident occurred is included. Thus, we account for potential individual, situational, and contextual influences on law enforcement responses to DV incidents that involves a child perpetrator and parent victim, known as CPV.

Method

Data and Participants

This study examines 1,113 calls for law enforcement service in a Midwest state wherein a child allegedly assaulted his or her parent.¹ The parent victim classification extends to individuals listed as a stepparent or grandparent of the alleged offender by the law enforcement official responding to the call for service. Based on data from the Office of Juvenile Justice and Delinquency Prevention (OJJDP) Report on Juvenile Offenders and Victims (Sickmund & Puzzanchera, 2014), approximately 10% of system involved youth (as victims or offenders) live in family structures with grandparents in the role of “pseudoparents.” For this reason, we also include grandparents as the parental figure in this study as an important consideration of CPV. To qualify for inclusion, offenders must be below 18 years of age. Similar to prior studies on CPV and related DV literature (Buzawa & Hotaling, 2006; Strom et al., 2014), incidents of CPV were based on the dyadic relationship involving one alleged offender and one victim. Strom et al. (2014) noted a significant methodological limitation when including multiple victims and offenders as complicating “the probabilities of arrest, because a single incident may involve multiple arrests, multiple victim/offender relationships, and varied victim and offender characteristics such as race, age or gender” (p. 434).

As shown in Table 1, the majority of offenders in this study were White, and 14 years of age or older. The parent victims also tended to be White. Slightly less than half of the victims sustained a documented injury from the child to parent violent incident. The most common gender pairing in these data was a female victim, male offender; female victim, female offender was second most common. The most common law enforcement response was arrest of the child perpetrator (55.5%), followed by informal resolution of the situation. As shown in Table 1, officers infrequently referred youth to a welfare or other agency (9.6%), most often opting for arrest or informal resolution (i.e., de-escalation) of the incident (34.9%).

Table 1.

Descriptive Characteristics of Child to Parent Domestic Violence Incidents.

Child to Parent Violent Incident Characteristics	Law Enforcement Response
Child to Parent Violent Incident Characteristics	Total (n = 1,113)	Informal (n = 388)	Referral (n = 107)	Arrest (n = 618)
Offender race (%)
White	71.1	69.8	77.6	70.7
Non-White	28.9	30.2	22.4	29.3
Offender age (% 14 and older)	76.1	74.2	77.6	77.0
Presence of victim sustained injury (% yes)**	46.3	41.2	35.5	51.3
Victim-offender gender dyad (%)
Female victim, male offender*	42.4	37.4	45.8	45.0
Female victim, female offender*	34.1	38.7	28.0	32.2
Male victim, male offender	16.1	38.7	20.6	16.3
Male victim, female offender (contrast group)	7.4	14.4	5.6	6.5
Female-headed households (Mean %)	5.9 (1.1)

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

Measures

Secondary data resulting from official police reports filed in response to the call for service comprise the data source for this variable. Law enforcement response at the incident level comprises the dependent variable in this study. Officer responses include arrest, referral to welfare or other social services agency, and informal handling of the incident (i.e., no-arrest).

Individual and situational characteristics

Incident level data extracted from the official police report provide individual and situational characteristics of the offender (i.e., the child), the victim (i.e., the parent), and the incident. Similar to prior research (e.g., Strom et al., 2014), variables included in this study are comprised of dichotomous indicators of offender race (White/non-White), offender age (below 14/14 and older), presence of victim injury (injury/no injury), and dyadic measures of the victim-offender gender pairing (Hong et al., 2012). The dichotomization of youth by age is informed by Kang and Lynch (2014) who find that decisions to call police in DV situations is heavily dependent upon the life stage of the individuals involved in the incident (older youth vs. children). These distinctions may also apply to police response to the incidents. Furthermore, in one of the few large national samples of reported CPV offenses, Walsh and Kreinert (2007) examined data from 2002 in 23 states, which included approximately 18,000 adolescents aged 21 and younger, who were reported for assaulting a parent or a stepparent. Overall, the most frequent offender profile was a White (76%) male (63%), between the ages of 14 and 17 (60%). For this reason, a dichotomous comparison of offenders younger than 14 years was made with older offenders. CPV is often described as a gendered crime, with research studies finding that CPV is committed by males toward their mothers (or stepmothers), particularly single mothers (Cottrell & Monk, 2004; Evan & Warren-Sohlberg, 1988; Ibabe & Jaureguizar, 2010; Walsh & Kreinert, 2007). For this reason, dyadic gender measures consist of same-sex pairings and cross-sex pairings (male victim, female offender; male victim, male offender; female victim, male offender; and female victim, female offender). The least likely gender pairing in CPV is a male victim-female offender dyad. This dyad forms the contrast (base) group in the analysis.

Contextual characteristics

We obtained publicly available data from the 2000 U.S. Census extracted at the county level, where CPV incident occurred to quantify the average level of family disruptions in that county. We measure family disruption as the percentage of the county households comprised of female-headed households (e.g., Warner & Pierce, 1993). A high percentage of female-headed households may create an atmosphere of ecological contamination that Schulenberg (2003) describes as having a potential influence on officer decision making, applied here to CPV.

Analytical Model

The categorical nature of the dependent variable, law enforcement response, calls for an analysis using a multinomial logistic regression model with robust standards errors (Long & Freese, 2017). Often described as an extension of logistic regression, multinomial regression allows for multiple dichotomous dependent variable responses in the same model used to compute the marginal effects of the selected covariate on the probabilities of choosing the alternatives. Results from this modeling approach include the logistic coefficient for the covariates (i.e., predictor variables) for each category of the dependent variable as compared with the reference category. Here, we include informal response (i.e., no-arrest, de-escalation only) as the reference category and make comparisons to the membership categories of referral to social welfare or other agency, and arrest. Although the use of the contextual variable family disruption would initially suggest a multilevel model, multinomial cluster model with robust standard errors produced results without substantive differences and therefore those results are described here.²

Results

Between group differences in the distribution of the covariates within the law enforcement response outcome variable was assessed with the chi-square statistic. Differences were found between outcome groups in the presence of victim injury variable, and in two of the four gender dyads, specifically, the dyads where the victim was female. Relative to an informal law enforcement response to incidents of CPV, the gendered nature of the victim-offender dyad, the presence of victim injury, the extent of family disruption in the surrounding community are influential in an officer’s decision to arrest the child perpetrator. As shown in Table 2, results indicate with the presence of a victim injury stemming from the CPV incident, officers were 1.61 times more likely to arrest the alleged offender than when no injury resulted from the incident. The gender of both the victim and offender also exhibited a statistically significant impact on the police officer’s decision to make an arrest. In comparison to the least common scenario in these data (i.e., a male victim-female offender dyad), officers were 1.95 times more likely to arrest a male aggressor (i.e., the son) when the victim was female (i.e., his mother). Interestingly, no individual or situational characteristics including injury or gender dyad were able to distinguish referrals to social welfare or other agencies for assistance from an informal law enforcement response.

Table 2.

Individual, Situational, and Contextual Predictors of Law Enforcement Response to Child–Parent Violence: Multinomial Regression Model With Robust Standard Error Model.

Individual, Situational, and Contextual Variables	Law Enforcement Response
	Referral to Social Welfare or Other Agency				Arrest
	RRR	95% Confidence Interval	Exp (B)	p > \|z\|	RRR	95% Confidence Interval	Exp (B)	p > \|z\|
Offender race (White = 1)	1.39	[.83, 2.31]	.33	.21	1.02	[.76, 1.35]	.02	.92
Offender age (>14)	1.25	[.74, 2.09]	.22	.40	1.20	[.89, 1.62]	.18	.24
Presence of injury	0.86	[.55, 1.35]	−.15	.50	1.61*	[1.24, 2.09]	.48	.00
Female victim-male offender dyad	2.16	[.85, 5.48]	.77	.10	1.96*	[1.19, 3.22]	.67	.01
Female victim-female offender dyad	1.25	[.48, 3.24]	.22	.65	1.26	[.76 2.07]	.23	.37
Male victim–male offender dyad	2.32	[.86, 6.3]	.84	.10	1.66	[.95, 2.90]	.51	.07
Female-headed household	.79*	[.65, .97]	−.23	.02	0.87*	[.78, .98]	−1.4	.02
Log likelihood = -1,001.94LR χ² (14) = 42.24, p = 0001Pseudo-R² = .021

Note. Response option contrast is no-arrest. Gender dyads contrasted to male victim-female offender pairing. The Institute for Digital Research and Education (2017) notes, “Logistic regression does not have an equivalent to the R-squared that is found in OLS regression; however, many people have tried to come up with one. There are a wide variety of pseudo-R-square statistics…. Because this statistic does not mean what R-square means in OLS regression (the proportion of variance for the response variable explained by the predictors), we suggest interpreting this statistic with great caution.” At the suggestion of a reviewer, we heed this warning regarding the pseudo r² and choose not to discuss it in text due to its potential inappropriate interpretation. RRR = relative risk ratio; LR = likelihood ratio; OLS = ordinary least squares.

Evidence consistently demonstrated that for officer decision making in CPV incidents, context of the incident matters. Relative to an informal (i.e., no-arrest) law enforcement response, incidents of CPV that occurred in jurisdictions with higher levels of family disruption were more likely to result in informal law enforcement responses relative to a decision to arrest the child or make a referral to a social welfare agency. For each percentage increase of female-headed households in the county, as compared with an informal response the officer was .79 times less likely to make an arrest and .87 times less likely to make referral to a social welfare or other agency. Stated differently, in areas that were more highly comprised of female-headed households police officers were significantly more likely to respond to the call for service and deescalate the incident, but take no further legal action. The odds of arrest occurring were significantly increased when an injury was present, and with female victim-male aggressor dyads holding all else constant.

Discussion

Through an examination of 1,113 incidents of CPV, we find factors influencing police officer decision making in cases of CPV demonstrates some similarity to the existing DV literature. Specifically, we find when victim injury occurs during a CPV incident, a more formalized response (i.e., arrest) is significantly more likely to occur. Whether law or jurisdictional policy guides this decision, the outcome of criminalizing the youth when they inflict an injury upon his or her parent during a physical altercation appears much more certain.

The nature of the relationship between victim and offender, which is gendered, is an important consideration when examining law enforcement response to CPV. Controlling for whether an injury is present, we find that male aggressors are more often criminalized through arrest when the victim is female (i.e., mother). Yet, male aggressors are no more likely than females to be arrested when they victimize a male (i.e., father). Similarly, the gender dyad when a female aggressor is involved does not appear to elevate the risk of either arrest or alternative intervention through referral. Why is it that sons who abuse their mothers are more likely to be arrested than sons who abuse their fathers or daughters who abuse either parent? This might be the outcome of decades of attention being directed at law enforcement response toward males’ use of violence against women in general (e.g., domestic and sexual violence).

Following Schulenberg’s (2003) theory of ecological contamination that purports context influences police officer decision to arrest youth, we find that context of family disruption decreases the likelihood of formal intervention, both arrest and alternative intervention. Whether the victim of CPV incidents in areas that exhibit higher levels of family disruption were not willing to cooperate in prosecuting their child, or whether the officer views these instances of CPV through a different lens is unknown. It is clear however that when victims of CPV do not receive additional support beyond incident de-escalation, the risk of re-victimization remains. Researchers must continue to determine how parent victims and their families can be best served when CPV occurs in the home.

Consistent with prior research, referral to welfare or other social services was the least likely law enforcement response to CPV incidents. Considering that police are often reluctant to offer assistance and/or do not view it as their “role,” this finding is not all that unexpected, but perhaps unfortunate (Russell & Light, 2006). Referral to services, welfare or other, might be the intervention most needed for a family in crisis.

Limitations

This study draws conclusions from the analysis of a relatively large data set of officially recorded calls for service. Although the data set was rather large (N = 1,113) and recent, it has limitations that must be noted. First, the usage of police records for studying CPV is problematic. CPV, like other forms of DV, is an underreported crime and it is plausible that when CPV is reported to the police, it is often because of the severity of the abuse (Cornell & Gelles, 1982). Resultantly less serious forms of CPV, but perhaps more common, are not captured in the data set.

Second, the data set did not contain several variables that would be beneficial to our understanding not only of CPV but law enforcement response to this unique type of violence. For instance, there was no information on historical police contact within the household or prior/ongoing involvement with child welfare system recorded in the data set. This is unfortunate, but not all that surprising since law enforcement, when responding to CPV incidents, often treat DV incidents as singular events, failing to recognize (and document) the interconnectedness between current and past events. Similarly, unknown is whether any incident that occurred within the year is a repeat call for service. Also lacking from the data set was information on single versus dual parent nature of the household within which the incident occurred.

Implications for Policy

To protect victims adequately while at the same time ensuring that child perpetrators are not unduly criminalized, training and education programs for law enforcement officers related to CPV is essential. It is unclear as to what extent law enforcement officers are trained and able to employ alternatives to a dichotomous arrest/no-arrest response in CPV situations. Miles and Condry (2016) examined 100 police case files of adolescent to parent violence along with 20 interviews with police officers finding that the majority of the cases were handled by uniformed officers who were not specially trained in responding to DV incidents. The use of discretion in defining DV situations as well as training of the officer responding to the call for service can have far-reaching, deleterious effects. Regarding the victim/parent, law enforcement response could affect future reporting behavior and the parental perceptions of appropriate use of police authority and/or perceptions of justice. As found here and noted elsewhere (see Miller, 2015), officer decisions may affect the support services the family receives. Miles and Condry note these effects may influence whether the child perpetrator desists from further violent behavior. They argue most parents want to “develop and maintain a non-violent relationship with their child” (p. 17) and would benefit from a police response that would allow parent victims to “report violence from their child to the police and access help without the consequence of criminalizing their child” (Miles & Condry, 2016, p. 821).

In conclusion, this study indicates that the gendered nature of CPV victim-offender dyads and the presence of victim injury influences police decision to arrest in lieu of de-escalation only. Also, officers are less likely to make referrals to social welfare agencies or make an arrest when a CPV incident occurs in neighborhoods with high levels of family disruption. In these instances, the officer works to deescalate the situation, but is significantly less likely to take any further action. Whether the victim is not willing to cooperate in prosecuting their child, or the officer views these instances of CPV through a different lens, researchers must continue to determine how parent victims and their families can be best served when CPV occurs in the home. Future studies should also continue to explore factors that influence law enforcement decision making and response to CPV.

Footnotes

Declaration of Conflicting Interests

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

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

Notes

Author Biographies

Gaylene S. Armstrong, PhD, is the director and professor in the School of Criminology and Criminal Justice at the University of Nebraska Omaha. Her research focuses broadly on criminal justice policy research. Her work has appeared in Justice Quarterly, Criminal Justice and Behavior, Crime and Delinquency, and the Journal of Research in Crime and Delinquency.

Lisa R. Muftic, PhD, is a chair and professor in the Department of Criminal Justice and Sociology at Western New England University. Her work explores the intersectionality of interpersonal violence perpetration, victimization, sex, and justice responses. Her work has appeared in journals such as Justice Quarterly, Crime and Delinquency, and the Journal of Research in Crime and Delinquency.

Leana A. Bouffard, PhD, is a professor and chair in the Department of Sociology at Iowa State University. Her research interests include life-course and developmental criminology, gender, and violence against women. Her work has appeared in Criminology, Justice Quarterly, Criminal Justice and Behavior, and Crime and Delinquency, among others.

References

Agnew

Huguley

(1989). Adolescent violence towards parents. Journal of Marriage and Family, 51, 699-711.

Buzawa

Hotaling

G. T.

(2006). The impact of relationship status, gender, and minor status in the police response to domestic assaults. Victims & Offenders, 1, 323-360.

Calvete

Orue

Gamez-Guadix

(2013). Child-to-parent violence: Emotional and behavioral predictors. Journal of Interpersonal Violence, 28, 755-772.

Calvete

Orue

Gamez-Guadix

Bushman

B. J.

(2015). Predictors of child-to-parent aggression: A 3-year longitudinal study. Developmental Psychology, 51, 663-676.

Calvete

Orue

Gamez-Guadix

Hoyo-Bilbao

de Arroyabe

E. L.

(2015). Child to parent violence: An exploratory study of the roles of family violence and parental discipline through the stories told by Spanish children and their parents. Violence and Victims, 30, 935-947.

Cohen-Filipic

Bentley

(2015). From every direction: Guilt, shame and blame among parents of adolescents with co-occurring challenges. Child and Adolescent Social Work Journal, 32, 443-454.

Condry

Miles

(2014a). Adolescent to parent violence: Framing and mapping a hidden problem. Criminology & Criminal Justice, 14, 257-275.

Condry

Miles

(2014b). Responding to adolescent to parent violence: Challenges for policy and practice. The British Journal of Criminology, 55, 1076-1095.

Contreras

Cano

(2014). Family profile of young offenders who abuse their parents: A comparison with general offenders and non-offenders. Journal of Family Violence, 29, 901-910.

10.

Contreras

Cano

(2015). Exploring psychological features in adolescents who assault their parents: A different profile of young offenders? The Journal of Forensic Psychiatry & Psychology, 26, 224-241.

11.

Cornell

C. P.

Gelles

R. J.

(1982). Adolescent to parent violence. Urban and Social Change Review, 15, 8-14.

12.

Correll

J. R.

Walker

S. C.

Edwards

T. C.

(2017). Parent perceptions of participating in a program for adolescents who are violent at home. Journal of Family Violence, 32, 243-255.

13.

Cottrell

Finlayson

(1996). Parent abuse: The abuse of parent by their teenage children (Report & Booklet). Ottawa, Ontario: The Family Violence Prevention Unit, Health Canada.

14.

Cottrell

Monk

(2004). Adolescent-to-parent abuse: A qualitative overview of common themes. Journal of Family Issues, 25, 1072-1095.

15.

Dichter

M. E.

Rhodes

K. V.

(2011). Intimate partner violence survivors’ unmet social service needs. Journal of Social Service Research, 37, 481-489.

16.

Erez

Tontodonato

(1989). Patterns of reported parent-child abuse and police response. Journal of Family Violence, 4, 143-159.

17.

Evans

E. D.

Warren-Sohlberg

(1988). A pattern analysis of adolescent abusive behavior toward parents. Journal of Adolescent Research, 3, 201-216.

18.

Felson

Cares

(2005). Gender and the seriousness of assaults on intimate partners and other victims. Journal of Marriage and Family, 67, 1182-1195.

19.

Gebo

(2007). A family affair: The juvenile court and family violence cases. Journal of Family Violence, 22, 501-509.

20.

Hamilton

Worthen

(2011). Sex disparities in arrest outcomes for domestic violence. Journal of Interpersonal Violence, 26, 1559-1578.

21.

Harbin

Madden

(1979). Battered parents: A new syndrome. American Journal of Psychiatry, 36, 1288-1291.

22.

Holt

(2016). Adolescent-to-parent abuse as a form of “domestic violence”: A conceptual review. Trauma, Violence & Abuse, 17, 490-499.

23.

Hong

Kral

M. J.

Espelage

D. L.

Allen-Meares

(2012). The social ecology of adolescent-initiated parent abuse: A review of the literature. Child Psychiatry & Human Development, 43, 431-454.

24.

Ibabe

Bentler

P. M.

(2016). The contribution of family relationships to child-to-parent violence. Journal of Family Violence, 31, 259-269.

25.

Ibabe

Jaureguizar

(2010). Child-to-parent violence: Profile of abusive adolescents and their families. Journal of Criminal Justice, 38, 616-624.

26.

Institute for Digital Research and Education (2017). Logistic Regression: SPSS Annotated Output. Retrieved from https://stats.idre.ucla.edu/spss/output/logistic-regression/ (accessed 8 August 2017).

27.

Kang

J. H.

Lynch

J. P.

(2014). Calling the police in instances of family violence: Effects of victim-offender relationship and life stages. Crime & Delinquency, 60, 34-59.

28.

Kennair

Mellor

(2007). Parent abuse: A review. Child Psychiatry & Human Development, 38, 203-219.

29.

Kennedy

T. D.

Edmonds

W. A.

Dann

K. T. J.

Burnett

K. F.

(2010). The clinical and adaptive features of young offenders with histories of child-parent violence. Journal of Family Violence, 25, 509-520.

30.

Kethineni

(2004). Youth-on-parent violence in a central Illinois county. Youth Violence and Juvenile Justice, 2, 374-394.

31.

Long

J. S.

Freese

(2017). Regression models for categorical dependent variables using Stata (2nd ed.). College Station, TX: Stata Press.

32.

Lyons

Bell

Frechette

Romano

(2015). Child-to-parent violence: Frequency and family correlates. Journal of Family Violence, 30, 729-742.

33.

Maas

C. J.

Hox

J. J.

(2005). Sufficient sample sizes for multilevel modeling. Methodology, 1, 86-92.

34.

Miles

Condry

(2016). Adolescent to parent violence: The police response to parents reporting violence from their children. Policing & Society, 26, 804-823.

35.

Miller

E. J.

(2015). Challenging police discretion. Howard Law Journal, 58, 521-556.

36.

Moulds

Day

Mildred

Miller

Casey

(2016). Adolescent violence toward parents—The known and unknowns. Australian & New Zealand Journal of Family Therapy, 37, 547-557.

37.

National Criminal Justice Reference Service. (2017). Special feature: Family violence. Retrieved from https://www.ncjrs.gov/family-violence/

38.

Pelletier

Coutu

(1992). Substance abuse and family violence in adolescents. Canada’s Mental Health, 40, 6-12.

39.

Police Executive Research Forum. (2015). Police improve response to domestic violence, but abuse often remains the “hidden crime.” Subject to Debate, 29(1), 1-3, 6-11.

40.

Rollwagen

Jacob

(2018). The victim-offender relationship and police charging decisions for juvenile delinquents: How does social distance moderate the impact of legal and extralegal factors? Youth Violence and Juvenile Justice, 16(4), 378-394.

41.

Routt

Anderson

(2011). Adolescent violence towards parents. Journal of Aggression, Maltreatment & Trauma, 20, 1-19.

42.

Russell

Light

(2006). Police and victim perspectives on empowerment of domestic violence victims. Police Quarterly, 9, 375-396.

43.

Schulenberg

(2003). The social context of police discretion with young offenders: An ecological analysis. Canadian Journal of Criminology and Criminal Justice, 45, 99-119.

44.

Sickmund

Puzzanchera

(2014). Juvenile offenders and victims: 2014 report. Pittsburg, PA: National Center for Juvenile Justice.

45.

Strom

Warner

Tichavsky

Zahn

(2014). Policing juveniles: Domestic violence arrest policies, gender, and police response to child-parent violence. Crime & Delinquency, 60, 427-450.

46.

Walsh

J. A.

Krienert

J. L.

(2007). Child-parent violence: An empirical analysis of offender, victim, and event characteristics in a national sample of reported incidents. Journal of Family Violence, 22, 563-574.

47.

Walsh

J. A.

Krienert

J. L.

(2009). A decade of child-initiated family violence: Comparative analysis of child-parent violence and parricide examining offender, victim and event characteristics in a national sample of reported incidents 1995-2005. Journal of Interpersonal Violence, 24, 1450-1477.

48.

Warner

B. D.

Pierce

G. L.

(1993). Reexamining social disorganization theory using calls to the police as a measure of crime. Criminology, 31, 493-517.

49.

Williams

Tuffin

Niland

(2017). “It’s like he just goes off, BOOM!”: Mothers and grandmothers make sense of child to parent violence. Child & Family Social Work, 22, 597-606.