POLICE RESPONSE TO DOMESTIC VIOLENCE: SITUATIONS INVOLVING VETERANS EXHIBITING SIGNS OF MENTAL ILLNESS*

• Date: 01 May 2015
• DOI: http://doi.org/10.1111/1745-9125.12067
• Published date: 01 May 2015

POLICE RESPONSE TO DOMESTIC VIOLENCE:

SITUATIONS INVOLVING VETERANS EXHIBITING

SIGNS OF MENTAL ILLNESS∗

FRED E. MARKOWITZ1and AMY C. WATSON2

1Department of Sociology, Northern Illinois University

2Jane Addams School of Social Work, University of Illinois at Chicago

KEYWORDS: mental illness, violence, veterans, police, military, stigma, attribution

theory

Drawing on attribution theory, research on police discretion, and public attitudes to-

ward mental illness, we examine attributional processes in police decision making in re-

sponse to domestic violence situations involving veterans and nonveterans with signs of

mental illness. Using data from experimental vignettes varying veteran status, victim in-

jury, and suspect compliance administered to a sample of 309 police officers, the results

indicate that 1) veterans are perceived as less responsible for troublesome behavior but

more dangerous than nonveterans, 2) suspects’ veteran status has a significant effect on

officers’ preference for mental health treatment versus arrest, and 3) part of the effect

of veteran status on officer response is mediated by internal and external attributions

for problematic behavior and by perceptions of dangerousness. The study empiri-

cally demonstrates countervailing processes in police decision making—recognition

of the causes for troublesome behavior and the need for mental health treatment

on the one hand and concern for community safety and enforcing the law on the

other.

Police officers are among those most likely to deal with persons with mental illness

in a variety of crisis situations (Engel and Silver, 2001; Lamb, Weinberger, and DeCuir,

2002). When police officers recognize that suspects have mental health problems, their

options include arresting the suspect, bringing them to hospitals for psychiatric treatment,

or resolving the issue informally (Teplin, 2000; Watson and Angell, 2007). The police

officer’s preference for more therapeutic (treatment) as opposed to punitive measures

(arrest) may be attributed to linking aggressive behavior to external causes beyond

individuals’ control (Corrigan et al., 2003). In this study, we examine attributional pro-

cesses as they apply to domestic violence involving persons with mental health problems.

Drawing from the literature on police discretion in domestic violence and mental health

crisis situations, as well as from theoretical approaches to attributions regarding mental

∗This research was supported in part by NIMH P20 MH085981. The authors are grateful to

Douglas Engelman and Anjali Fulambarker for their research assistance. We also appreciate the

very helpful comments provided by Richard Felson and Kirk Miller. Direct correspondence to Fred

E. Markowitz, Department of Sociology, Northern Illinois University, DeKalb, IL 60115 (e-mail:

fredm@niu.edu).

C2015 American Society of Criminology doi: 10.1111/1745-9125.12067

CRIMINOLOGY Volume 53 Number 2 231–252 2015 231

232 MARKOWITZ & WATSON

illness, we use a set of experimental vignettes administered to police officers to test

a model of the extent to which attributions for the causes of troublesome behavior

mediate the effects of military veteran status, injury, and noncompliance on the likeli-

hood of arrest, mental health treatment, and informal resolution in domestic violence

situations.

POLICE DISCRETION IN DOMESTIC VIOLENCE AND MENTAL

HEALTH CRISES

Despite conflicting findings about whether arrest deters domestic violence, many

states’ laws and police agency policies either mandate or “prefer” arrest given probable

cause (Philips and Sobol, 2010; Zeoli, Norris, and Brenner, 2011). Although the inten-

tion of these policies is to limit police discretion and favor arrest, police still exercise

considerable discretion over whether to make arrests or handle matters informally.

According to data from the National Incident-Based Reporting System, arrests are made

in fewer than half of violent incidents involving domestic partners (Eitle, 2005; Hirschel

et al., 2007, 2008). Although arrests are somewhat more likely in states with mandatory

or “preferred” arrest policies, police decisions regarding whether to make arrests are

complex. Previous research, using data from police reports, self-reports from officers, and

experimental vignettes has shown that when 1) offenses involve weapons, 2) more severe

injuries, 3) are repeat incidents, and 4) additional witnesses are present, officers are more

likely to make an arrest (Bachman and Coker, 1995; Dichter et al., 2011; Eigenberg,

Scarborough, and Kappeler, 1996; Eitle, 2005; McLaughry et al., 2013; Robinson and

Chandek, 2000). Also, when suspects are not compliant—especially in terms of verbally

expressed hostility—officers are more likely to make an arrest (Novak and Engel, 2005;

Worden and Shepard, 1996).

Only one study specifically examined the role of suspects’ mental illness in arrest

and referral for treatment decisions in domestic violence. Finn and Stalans (2002) used

experimental scripts that varied suspects’ mental state (hallucinating, drunk, or “nor-

mal”), antagonism or cooperativeness between disputants, and presence or absence of

victim injury. Their findings indicated that when officers believed the suspect presented

a significant risk to the victim if left in the home, they were more likely to arrest them

(based on responses coded from open-ended interviews). Also, officers were more likely

to indicate they would refer mentally ill suspects for civil commitment when they believed

that psychiatric hospitals would admit violent patients and when they thought they were

dangerous.

Although their training often includes managing mental health crises, police, who view

troublesome situations as “law enforcers,” are motivated to maintain their authority in

conflict situations, often invoking the power of arrest to do so (Watson and Angell, 2007).

One study of police encounters with persons with mental illness showed that whether

suspects are under the influence of drugs or are noncompliant with officers, as well as the

seriousness of their offense, predicted the likelihood of arrest (Engel and Silver, 2001).

Persons with mental illnesses also are more likely to be resistant in police encounters,

thus increasing the chances that police will use force and arrest (Johnson, 2011; Novak

and Engel, 2005).