Intimate Partner Violence and Animal Abuse in an Immigrant-Rich Sample of Mother–Child Dyads Recruited From Domestic Violence Programs

Abstract

We examined rates of animal abuse in pet-owning families experiencing intimate partner violence (IPV). We also examined whether higher levels of IPV (as measured by subscales from the Conflict Tactics Scales) predicted increased risk for partner-perpetrated animal abuse. Our sample included 291 mother–child dyads, where the mothers sought services from domestic violence agencies. Nearly half the sample is comprised of Mexican immigrants. Mothers reported that 11.7% of partners threatened to harm a pet and 26.1% actually harmed a pet, the latter of which represents a lower rate than in similar studies. When examining animal abuse by “Hispanic status,” follow-up analyses revealed significant omnibus differences between groups, in that non-Hispanic U.S.-born partners (mostly White) displayed higher rates of harming pets (41%) than either U.S.-born or Mexican-born Hispanic groups (27% and 12.5%, respectively). Differences in rates for only threatening (but not harming) pets were not significant, possibly due to a small number of partners (n = 32) in this group. When examining whether partners’ IPV predicted only threatening to harm pets, no IPV subscale variables (Physical Assault, Psychological Aggression, Injury, or Sexual Coercion) were significant after controlling for income, education, and Hispanic status. When examining actual harm to pets, more Psychological Aggression and less Physical Assault significantly predicted slightly higher risk of harm. However, Mexican-born partners had nearly 4 times lower risk of harming a pet. Overall, these results suggest that Hispanic men who are perpetrators of IPV are less likely to harm pets than non-Hispanic perpetrators of IPV, particularly if Mexican-born. Considering that the United States has a significant proportion of Mexican immigrants, it may be worthwhile to explore the topics of IPV and animal abuse within this group.

Keywords

domestic violence intimate partner violence animal cruelty pet abuse Mexican immigrants and intimate partner violence

Introduction

The link between family violence and animal abuse has become an increasingly examined area of research in the past two decades (e.g., Ascione, 1997; Ascione et al., 2007; Faver & Strand, 2003; Flynn, 2011; Volant, Johnson, Gullone, & Coleman, 2008). Studies examining animal abuse in families experiencing intimate partner violence (IPV) have found that men who abuse their partners also abuse animals at rates significantly higher than men in nonabusive adult relationships (Ascione et al., 2007; Volant et al., 2008) and that abuse of family pets is a risk factor for men perpetrating IPV (Walton-Moss, Manganello, Frye, & Campbell, 2005). Another study has suggested that abusive partners who also abuse pets display more forms of violence and greater use of controlling behaviors with their partners (Simmons & Lehmann, 2007). In addition, abusers may use pets as leverage to control their partners. For example, one study found that abusers had used threats of harm to pets as a way to coerce their partners into committing crimes (Loring & Bolden-Hines, 2004). Finally, concern for a pet’s well-being can influence a victim’s decision to leave her partner (Ascione et al., 2007) and make it more difficult for her to leave.

Animal abuse within families experiencing IPV also has a significant impact on children. Numerous studies have found that children living in families experiencing IPV are more likely to engage in animal abuse than children from nonviolent homes (e.g., Becker, Stuewig, Herrera, & McCloskey, 2004; Currie, 2006; DeGue & DiLillo, 2009; Volant et al., 2008). Moreover, some studies suggest that both children and adolescents frequently rely on pets as a way of managing stress (e.g., Melson, Schwarz, & Beck, 1997).

Overall, understanding the role pets play in families with IPV, and the relation between IPV and animal abuse, are important areas for exploration among researchers and clinicians.

Rates and Prediction of Pet and Animal Abuse

When studying families experiencing IPV, it is important to examine rates of threatening to harm pets as well as actually harming them as these are different behaviors with potentially different predictors and outcomes. Several small studies, with sample sizes ranging from 28 to 43, interviewed pet-owning women who were experiencing IPV and sought services from domestic violence (DV) shelters. In these studies, 14% to 21% of women reported that a partner only threatened to harm a pet (Ascione, 1997; Carlisle-Frank, Frank, & Nielsen, 2004; Faver & Strand, 2003; Flynn, 2000). The rates for actual harm to pets ranged from 46% to 57%, with the exception of one study that reported 26% (Flynn, 2000). In a somewhat larger study (n = 102), Volant et al. (2008) interviewed pet-owning women in DV shelters and found rates comparable to those of the other studies, where 12% of partners threatened to harm pets and 53% actually harmed a pet. In a comparison group of women not experiencing IPV, threat and harm rates were 6% and 0%, respectively.

Finally, Ascione et al. (2007) conducted a similar study of 100 women but included a more extensive battery of instruments to assess IPV and treatment of animals among family members. Their estimates of partners only threatening to harm animals (18%) and actually harming them (54%) were also consistent with previous studies (comparison group rates were 12.5% and 5%, respectively). When they conducted further analyses to predict threats and harm to pets, they found that scores on the Conflict Tactics Scales (CTS; Straus, Hamby, Boney-McCoy, & Sugarman, 1996) were the strongest predictor of both threats and harm, above and beyond demographic factors such as education or ethnicity. Moreover, they found that threat to pets was predicted by CTS Minor Physical Violence and Verbal Aggression subscale scores and that actual harm to or killing of pets was predicted by Severe Physical Violence subscale scores. However, as far as we are aware, this study is the only one that has examined types of IPV as predictors of threat and harm to pets.

When examining animal abuse rates in the general population, rather than in a clinical population of women seeking DV services, Vaughn et al. (2009) utilized the large (n = 43,093) representative National Epidemiologic Survey on Alcohol and Related Conditions (NESARC) sample and found that the lifetime prevalence of animal cruelty among U.S. adults was 1.8%. Although this study focused on general animal cruelty rather than pet abuse per se, and relied upon only one item to assess animal cruelty, they did yield some interesting findings with regard to ethnicity: African Americans, Native Americans/Asians, and native-born Americans (i.e., those born in the United States) reported comparatively high levels of cruelty to animals, whereas Hispanics reported comparatively low levels of such behavior.

Thus, in families experiencing IPV, partner animal abuse rates may vary depending not only on the nature and severity of IPV perpetrated by the partner but also on ethnicity and/or whether or not the partner was born in the United States.

Animal Abuse and IPV in Hispanic Populations

The above-cited studies of women experiencing IPV were comprised mostly of non-Hispanic participants. There is still little known about animal abuse—and its relation to IPV—among Hispanics, including Latino immigrants. Reporting of and attitudes toward IPV among Hispanic populations differ from those of Whites (see Cho, Velez-Ortiz, & Parra-Cardona, 2014); thus, it’s possible that ethnic group differences exist for animal abuse as well. Some suggest that Hispanics own fewer pets than Whites (Pew Research Center, 2006) but that pet ownership is more common among battered women (Faver & Cavazos, 2007) and may provide an important source of emotional support to these women (Flynn, 2000; Risley-Curtiss, Holley, & Wolf, 2006). Abusers may use a partner’s attachment to a pet as a way to control them, but there is mixed evidence for whether Hispanics or Mexican immigrants share the same view toward (or attachment to) pets as White Americans (e.g., Risley-Curtiss et al., 2006; Siegel, 1995).

To our knowledge, only one study has examined the issue of animal abuse in Hispanic families experiencing IPV. Faver and Cavazos (2007) looked at animal abuse rates in a sample of 151 women in South Texas DV programs. In this sample, which was 74% Hispanic, 36% of the women reported that their partners had threatened, harmed, or killed their pets. This estimate is lower than found in the previously cited studies with mostly non-Hispanic participants. When examining ethnic groups separately, they found that 32% of the Hispanic women reported that their partners abused their pets, compared with 52% of White non-Hispanic women. This difference was not significant, but it does suggest that animal abuse rates may be lower in Hispanic families experiencing IPV. The Faver and Cavazos study did not specify whether or not the Hispanic participants were U.S.-born. While Vaughn et al. (2009) found lower rates of animal cruelty in Hispanics and higher rates in U.S.-born participants, no one has examined rates and predictors of animal abuse in samples with a high proportion of Hispanic immigrants.

Overall, considering that the United States has a significant and growing number of Hispanic and Mexican immigrant populations, it’s important to examine animal abuse in these populations, especially in families experiencing IPV, to effectively provide services to these women, their children, and their pets.

This Study

This study examines rates of pet abuse (partner-perpetrated threats as well as actual harm) in a sample of mothers experiencing IPV, where our sample contains a high proportion of Hispanic participants, particularly Mexican immigrants. It also examines the relation between different types of IPV (Physical Assault, Psychological Aggression, Injury, and Sexual Coercion) and partner-perpetrated threats/harm.

To accomplish this, we present data from the first wave of a longitudinal study of 291 mothers and their 7- to 12-year-old children who resided at DV shelters or received nonresidential services from a DV agency, and who reported owning a pet animal currently or within the past year. The overall longitudinal study serves as a continuation and expansion of the work of Ascione et al. (2007) and includes a larger and more ethnically diverse sample as well as a breadth of instruments to assess IPV, animal abuse, and behavior.

For the current study, due to the composition of our sample (high proportion of Hispanic participants), we expect somewhat lower rates of animal abuse compared with that of similar samples of women seeking DV services. We also expect that higher levels of IPV will predict both increased threats to pets as well as increased actual harm to pets; specifically, we expect that more Psychological Aggression will predict greater likelihood of threats, whereas more Physical Assault will predict greater likelihood of actual harm to pets.

Method

Recruitment

Recruitment of participants occurred at 22 DV agencies between March 2011 and June 2014. Inclusion criteria for the study were as follows: (a) a woman has experienced IPV within the past 12 months, (b) has at least one child aged 7 to 12 years living in her home, and (c) mother/child have or have had at least one pet animal in their residence in the past 12 months.

In addition to posting flyers describing the project, shelter staff also approached women with children within the age range targeted in this study and described the purpose of the project, the need to focus on women and children who have or have had pet animals in the past year, the interviews and assessments that would be used, measures to ensure the confidentiality of the data collected, the potential for continuing participation after 18 months, and remuneration for participation.

A shelter or outreach program staff member was designated as the project contact person at each site. Both in these initial contacts and in the informed consent and child assent protocols, staff made it clear that participation is entirely voluntary and would not affect the level and quality of services provided to IPV victims and their children.

We selected DV sites for conducting assessments for a number of reasons. These sites, by their nature, are secure locations and provide greater safety for our participants than would our university location. Travel to the university might be, for some participants, an atypical journey and could compromise safety planning. Furthermore, should emotional distress occur during assessments, it would be easier and more effective to provide care at DV sites, where trained counselors are available. Finally, we expected that DV Program sites would provide a greater sense of familiarity and comfort to participants than would our university setting.

Informed Consent

Informed consent for mother and child was obtained from participant mothers, and assent from participant children. Each mother and child was separately given a description of the study in layperson’s language and was told the general focus of the assessments. Each participant was also informed that his or her responses during assessment would not be shared with the other member of the dyad and that all responses during assessment would be confidential. Participants were informed that confidentiality would only be breached if previously unreported maltreatment was disclosed by either member of the dyad or if intended harm to self or others was described during the course of assessments.

Interviews

Designated staff from each agency were trained to recruit participants, obtain voluntary consent and assent, and administer standardized surveys to eligible women and children. The training also included coding instructions and safety strategies to protect participants’ privacy and confidentiality. Staff members interviewed the mother first, and then met individually with the child to administer the child interview. Mothers received US$65 for the interview, plus an additional US$15 to be given to the child. Interviews generally took no more than 2 hr for the mother and 1 hr for the child.

Assessments

Once enrolled in the study, participants were administered a battery of measures to assess demographics, IPV, animal and pet abuse among family members, and a variety of child behaviors, including empathy, attachment to pets, and behavior problems. All assessments were translated to Spanish as well, to interview Spanish-speaking participants. Assessments used in the current study included the following:

Demographics

Demographic data, filled out by the mother, included age, race/ethnicity, education level, birth country, and primary language spoken for mother, child, and partner. It also included household income, age, and sex of other children living in the home, number of years having experienced IPV, and the relationship with the abusive partner (e.g., married, separated, etc.). Finally, it included the number and type of pets in the home, as well as the age of the pets and how long they’ve had them.

CTS2

The CTS2 (Straus et al., 1996) was used to assess IPV, where the mother was asked to report on her partner’s behavior in the past year. The CTS2 has been used extensively in research on IPV. Response choices for each item range from 0 (This has never happened to me) to 6 (More than 20 times in the past year), as well as 7 (Not in the past year, but it did happen before). The CTS2 yields a total score and five subscale scores (Physical Assault, Psychological Aggression, Injury, Sexual Coercion, and Negotiation) all of which have reported internal consistencies of α = .79 or higher. We used the midpoint scoring method recommended by Straus et al. (1996). For example, if a woman indicated that her partner’s behavior occurred 11 to 20 times during the past year, this item would receive a score of 15.

When examining mothers’ ratings on the CTS2, excluding Negotiation items (which are positive in nature), the mean total frequency score for the sample was 257 (SD = 217). In other words, the mothers experienced an estimated average of 257 abusive events in the year that preceded the interview (out of a possible 825). When examining the CTS2 abuse subscales, scores for the estimated total frequency of abusive acts averaged 108 for Psychological Aggression (SD = 75, max 200), 87 for Physical Assault (SD = 90, max 300), 22 for abuse resulting in Injury (SD = 31, max 150), and 39 for Sexual Coercion (SD = 55, max 175).

Pet Treatment Survey (PTS)

The PTS (Ascione, 2011) is a structured interview that asks about treatment of pets in the home, including how pets are treated by the mother, the child in the study, other children in the home, and the mother’s partner. It includes qualitative information about threats and actual harm to pets, care of pets, and whether pets influenced the mother’s decision process about leaving the abusive partner. The PTS is a revision of the Battered Partner Treatment Survey—Pet Maltreatment Assessment (BPSS; Ascione & Weber, 1995), which was used in two separate studies examining animal abuse in the context of IPV (Ascione et al., 2007; Volant et al., 2008).

Cruelty to Animals Inventory (CAI)

This measure (Dadds et al., 2004) assesses animal abuse perpetrated by children. It includes both caregiver report (CAI-P) and child self-report (CAI-C) forms. The CAI was developed for use with 6- to 13-year-olds and has excellent psychometric properties as reported by the instrument developers (Dadds et al., 2004). Coefficient α, with zero scores deleted, was .88 for the CAI-P and .89 for the CAI-C. Correlations between scores on the CAI-P and CAI-C, with zero scores deleted, were reported to be positive and moderate in size (for boys, r = .32 and for girls, r = .43).

Analysis

We conducted logistic regression analysis to examine our hypothesis that higher levels of IPV would predict increased likelihood of partner-perpetrated animal abuse. Specifically, we examined whether mean scores on the Physical Assault, Psychological Aggression, Injury, and Sexual Coercion subscales of the CTS2 significantly predicted whether the partner (a) only threatened to harm or kill a pet, and (b) actually harmed or killed a pet. We included annual income and partner’s highest education completed as covariates.

Results

Demographic Information

We recruited and interviewed 291 mother–child dyads, with the goal of obtaining 300. Mothers’ age ranged from 21 to 65, with a mean of 36.3 (SD = 7.43). Children were required to be between 7 and 12 years for inclusion in the study, and had a mean age of 9.07 (SD = 1.6). Of the children, 52.6% were boys and 47.4% were girls. Table 1 shows the ethnic breakdown for the sample.

Table 1.

Ethnic Breakdown for Mother, Child, and Partner.

	Mother	Child	Partner
White	26.9%	22.0%	32.5%
Hispanic	60.7%	55.3%	56.4%
Black	3.4%	3.4%	4.8%
Pacific Islander	0.3%	—	0.3%
Asian	0.3%	0.3%	0.7%
American Indian/Alaskan	1.7%	1.0%	1.0%
Mixed race	6.6%	17.9%	4.2%

Of the 291 mothers, 54.6% were born outside the United States (48.1% were born in Mexico). Of the children, 274 (94%) were born in the United States. Thirty-four percent of participants were interviewed in Spanish and filled out assessments that had been translated into Spanish.

For employment, 35.6% of mothers were employed full-time, 10% part-time, 29.8% were stay-at-home parents, 24.2% unemployed, and 0.3% retired. For current annual household income, the largest percentage (26%) fell into the US$20K to US$29K bracket, 37.8% fell below that level, and the remaining 36.2% made more than US$30K annually. Twenty percent had an education level of sixth grade or less, 16% had seventh to 11th grade, 39% were high school graduates, while the remaining 25% had some college or more.

Partner data: Mothers reported that their partners’ age ranged from 21 to 72 years, with a mean of 38.6 (SD = 8.66). Fifty-one percent were born in the United States, while 43.6% were born in Mexico. For employment, 71.6% were employed full-time, 3.2% part-time, 23% unemployed, 1.4% retired. Twenty-two percent had a sixth-grade education or less, 14.9% had seventh to 11th grade, 47.2% were high school graduates, and 16% had some college or a college degree.

Mothers also reported on the status of their relationship with their partners at time of interview: 29.2% were married and living with their partner, 29.2% married but not living together, 16.3% split up with partner but previously lived together, 8.7% divorced, 5.2% unmarried and living with partner, 4.2% legally separated, 1.7% dating but not living with partner, 1% split up with partner they never lived with, and 3.8% other status. Mothers reported that 73.4% of the children in the study were the biological child of the partner.

In terms of experiences with IPV, mothers reported having experienced IPV for a mean of 9.03 years (SD = 6.7), and 72% reported that the visit in which they were recruited for the study was the first time they had sought help regarding IPV.

Pets and Animal Abuse

At the time of interview, 90% of participants had a pet in the home, while the remaining 10% had a pet within the last 12 months (as required for study inclusion). Fifty-nine percent of the sample had one pet, 22% two pets, and 19% had three or more pets. Seventy-three percent had dogs, 35% cats, 8% birds, and 14% other kinds of pets.

Mothers reported that 11.7% of partners threatened to harm (but did not harm) a family pet, whereas 26.1% of partners actually harmed or killed a pet. When asked if Mom’s partner had harmed or killed a pet, 26.2% of children reported that he had.

In addition, 18.8% of mothers reported that concern over a pet’s welfare prevented her from coming to a DV shelter sooner, while 11.1% reported that concern over a pet’s welfare affected her decision-making about whether to stay with or leave a partner.

Because our rates of animal abuse are lower than those of other studies of families experiencing IPV, particularly when utilizing similar measures of animal abuse (i.e., threat and harm), we acknowledged that the composition of our sample may have contributed to these results. Our sample has a significant proportion of Hispanic and immigrant participants: 56.4% of the mothers’ partners are Hispanic and 43.6% are Mexican-born. And because attitudes toward pet animals may vary substantially depending on cultural factors such as ethnicity and country of origin, we examined threats and harm to pets separately for three categories: non-Hispanic U.S.-born, Hispanic U.S.-born, and Hispanic Mexico-born. Our sample also has seven Hispanic partners born in other countries (e.g., Guatemala) and six non-Hispanic partners born outside the United States (e.g., Europe); however, because it is too difficult to draw conclusions from samples this small, these latter groups were excluded from follow-up analyses. Likewise, because the non-Hispanic participants were primarily White, all non-Hispanics were considered a single group. Finally, those who had missing data for ethnicity or place of birth were excluded.

Table 2 shows that when it comes to only threatening harm to pets, Hispanic U.S.-born men showed the highest rates. However, omnibus differences in threat rates were not significant, χ²(2) 4.1, p = .129. For actually harming/killing a pet, the non-Hispanic U.S.-born men displayed higher rates and Mexican-born men the lowest; the omnibus group differences were significant, χ²(2) 23.6, p < .001. These estimates suggest that Hispanic men who are perpetrators of IPV are less likely to harm pets than non-Hispanic perpetrators of IPV and that Mexican-born perpetrators of IPV are even less likely to actually harm a pet.

Table 2.

Comparing Partner Pet Abuse Rates by Ethnicity/Place of Birth (n = 272).

	Non-Hispanic U.S.-Born (n = 108)	Hispanic U.S.-Born (n = 37)	Hispanic Mexico-Born (n = 127)
Partner only threatened to harm pet	12 (11.1%)	8 (21.6%)	12 (9.5%)
Partner actually harmed/killed pet	43 (41.0%)	10 (27.0%)	15 (12.5%)

Using IPV to Predict Threat and Harm to Pets

When examining whether mothers’ reports of their partners’ IPV predicted threats to harm pets, we found that higher Psychological Aggression scores slightly but significantly increased the likelihood of threats, above and beyond the effects of income, partner’s education, and other types of abuse (Table 3). This is best shown by examining the odds ratio, referred to as Exp(B) in our table, where an estimate significantly greater than 1.0 indicates increased risk. Injury, sexual coercion, income, and education were not significant predictors.

Table 3.

Predicting Threats and Actual Harm to Pets.

						95% CI for Exp(B)
	B	SE	df	p	Exp(B)	Lower	Upper
Threats to Harm Pet
IPV (Physical)	−.038	.053	1	.472	0.963	0.868	1.068
IPV (Psych)	.070	.034	1	.042	1.072	1.003	1.146
IPV (Injury)	.048	.053	1	.365	1.049	0.945	1.165
IPV (Sexual)	.034	.029	1	.245	1.035	0.977	1.096
Income	−.189	.113	1	.095	0.828	0.663	1.033
Partner’s education	.027	.068	1	.692	1.027	0.900	1.173
Actual harm to pet
IPV (Physical)	−.120	.045	1	.007	0.887	0.812	0.968
IPV (Psych)	.063	.024	1	.009	1.065	1.016	1.117
IPV (Injury)	.099	.051	1	.054	1.104	0.998	1.221
IPV (Sexual)	−.055	.028	1	.054	0.947	0.895	1.001
Income	−.009	.075	1	.900	0.991	0.855	1.148
Partner’s education	.196	.058	1	.001	1.216	1.086	1.362

Note. CI = confidence interval; IPV = intimate partner violence. The bold is for significant results.

For actual harm to pets, we found that Physical Assault, Psychological Aggression, and partner’s education significantly predicted increased likelihood of the partner harming/killing a pet. However, the Physical Aggression and partner’s education estimates were in the opposite direction that previous research has indicated, in that more physical violence predicted decreased likelihood of harming pets and more education increased the likelihood of harming pets.

Given that some of the above results weren’t consistent with previous studies (or our predictions), and given the differences in harm rates depending on ethnicity and country of origin, we conducted a series of follow-up regressions to account for whether a partner was non-Hispanic U.S.-born, Hispanic U.S.-born, or Hispanic Mexico-born (referred to from now on as “Hispanic status”).

Results (Table 4) showed that when including partner’s Hispanic status, none of the predictor variables remained significant when predicting threats. When predicting actual harm to pets, Physical Assault and Psychological Aggression remained significant but education did not. Instead, being Mexican-born was a significant predictor and the strongest predictor of all. In other words, more Psychological Aggression predicted significantly increased likelihood of harming/killing a pet, whereas higher rates of Physical Assault and being Hispanic Mexican-born significantly predicted lower likelihood of harming a pet.

Table 4.

Predicting Threats/Actual Harm, Including “Hispanic Status” Variable.

						95% CI for Exp(B)
	B	SE	df	p	Exp(B)	Lower	Upper
Threats to Harm Pet
IPV (Physical)	−.033	.054	1	.545	0.968	0.870	1.076
IPV (Psych)	.062	.036	1	.085	1.064	0.991	1.143
IPV (Injury)	.046	.054	1	.402	1.047	0.941	1.164
IPV (Sexual)	.035	.030	1	.239	1.036	0.977	1.099
Income	−.119	.116	1	.306	0.888	0.707	1.115
Partner’s education	−.033	.091	1	.717	0.967	0.809	1.157
Hispanic, U.S.-born	.797	.534	1	.135	2.220	0.780	6.319
Hispanic, Mexican-born	−.381	.616	1	.537	0.683	0.204	2.286
Actual harm to pet
IPV (Physical)	−.109	.046	1	.019	0.897	0.819	0.982
IPV (Psych)	.067	.026	1	.011	1.070	1.016	1.126
IPV (Injury)	.087	.053	1	.102	1.090	0.983	1.209
IPV (Sexual)	−.051	.029	1	.077	0.950	0.897	1.006
Income	−.040	.080	1	.615	1.041	0.890	1.218
Partner’s education	−.096	.072	1	.184	1.101	0.955	1.269
Hispanic, U.S.-born	−.493	.436	1	.258	0.611	0.260	1.436
Hispanic, Mexican-born	−1.361	.452	1	.003	0.256	0.106	0.621

Note. CI = confidence interval; IPV = intimate partner violence. Bold is for significant results.

Discussion

We examined IPV and animal abuse in a sample of 291 mother–child pairs who sought DV services. Mothers reported that 11.7% of partners threatened to harm (but did not harm) a family pet, whereas 26% of partners actually harmed or killed a pet. The threat rates are consistent with, albeit on the low end of, previous studies of women in DV programs, which range from 12% to 21% (Ascione, 1997; Ascione et al., 2007; Carlisle-Frank et al., 2004; Faver & Strand, 2003; Flynn, 2000; Volant et al., 2008). However, the rate for actually harming/killing a pet (26%) is considerably lower than the 46% to 57% reported in all but one of these other studies.

When we examined these results further and looked at threat and harm rates by “Hispanic status” (i.e., whether a partner was non-Hispanic U.S.-born, Hispanic U.S.-born, or Hispanic Mexico-born), we found that threat rates showed no significant differences. The sample size of men who only threatened (but did not actually harm) a pet was small (n = 32), possibly contributing to this result. However, rates of actual harm did show significant omnibus differences among the three groups, where non-Hispanic U.S.-born men had the highest harm rates and Hispanic Mexico-born men had the lowest.

We also examined whether Physical Assault, Psychological Aggression, Injury, and Sexual Coercion subscales (from the CTS; Straus et al., 1996), along with income and partner education, predicted threat and/or harm to pets. Higher Psychological Aggression slightly increased the likelihood of threatening harm to pets. However, when we included the partners’ Hispanic status in the model, Psychological Aggression did not achieve significance. A larger sample may have withstood the addition of the Hispanic status variable (and the resulting loss of two degrees of freedom), yielding a significant result. Yet the odds ratio for Psychological Aggression was not much greater than 1.0, suggesting that it is not a strong predictor of harming pets, at least in this sample. However, given that Ascione et al. (2007) found that a “Verbal Aggression” scale significantly predicted threats to pets, it may be worth further study to examine the value of psychological abuse as a unique predictor of threatening a partner with pet harm.

For predicting actual harm to pets, Physical Assault, Psychological Aggression, and partner’s education were significant. However, the Physical Assault and partner’s education estimates were in the opposite direction that previous research has indicated, in that less physical violence and higher education indicated slightly increased likelihood of harming/killing pets. These unexpected results make possible sense in our particular sample, for two reasons: (a) the Mexican-born partners, who showed significantly lower levels of harm to pets, often had no more than a sixth-grade education, and (b) Mexican-born partners displayed significantly higher rates of IPV than did the U.S.-born groups (data not shown). Yet, once we included Hispanic status, the Psychological Aggression variable remained significant, Physical Assault remained significant with the same unexpected direction of effect, education was no longer significant, and being Mexican-born was strongly significant.

In other words, even when accounting for partners’ ethnicity and place of birth, higher Psychological Aggression significantly predicted a slightly increased likelihood of actually harming/killing a pet. Higher Physical Assault predicted marginally lower likelihood of harm to pets. Both of these results are inconsistent with Ascione et al.’s (2007) work, which examined a different version of the CTS and found that Minor Physical Violence and Verbal Aggression predicted threat to pets, whereas Severe Physical Violence predicted harming/killing pets. Overall, our results for Psychological Aggression and Physical Assault were modest, while Injury and Sexual Coercion subscales showed no predictive power, suggesting that the first two may have some value in predicting animal harm. However, only future studies, ideally with larger and more representative samples, can confirm this.

Finally, our strongest finding was that Mexican-born men were nearly 4 times less likely to harm an animal than non-Hispanic U.S.-born men. This is consistent with the findings of Vaughn et al. (2009), who examined animal cruelty in the large, nationally representative NESARC sample and found that native-born Americans showed significantly higher prevalence of animal cruelty, while Hispanics reported significantly lower prevalence of animal cruelty.

There are many possible explanations for the differences between our results and those of other studies, including the Ascione et al. (2007) study. One is underreporting: At least one study has shown that Hispanic women may be prone to underreport sexual abuse and IPV for a variety of reasons (Ahrens, Isas, Rios-Mandel, & Lopez, 2010). The women in our sample did report significant levels of IPV from their partners, but it is possible they were not comfortable reporting on other problems, including the abuse of family pets. Another concern is that some newly immigrated or undocumented Hispanic women may report IPV because they’re seeking DV services, but may not feel comfortable reporting other behaviors that they believe could run the risk of attracting police or leading to deportation.

Another possible reason that we show significantly lower rates of pet harm among Hispanics and particularly Mexican-born Hispanics is that these groups, being from different cultures, may have different attitudes toward animals and pets than do Americans. For example, if a pet is not considered a family member, then an abuser may be less likely to harm it as a method of coercing or punishing a partner. However, this idea does not hold up well in the literature, and Hispanics regard pets as family members on a level similar to that of non-Hispanics (see Faver & Cavazos, 2007; Risley-Curtiss et al., 2006). Finally, it is possible that American and Mexican cultures differ in their views toward harming women versus animals. Only further study can examine these questions more thoroughly.

This study did have some limitations. First, it lacked a comparison group. Although similar studies had comparisons groups (Ascione et al., 2007; Volant et al., 2008) and showed that animal abuse rates are significantly higher in those families experiencing IPV, a comparison group of Mexican-born families without IPV would have been very informative. Second, we examined women and children who sought DV services, and this population may not represent other families experiencing IPV in terms of their rates of pet abuse. Finally, while we did collect data on where the mothers, children, and partners were born, we do not know when those who immigrated actually arrived in the United States and thus cannot estimate their degree of acculturation, something that may influence results. We also do not know if their residence in the United States is documented or undocumented, which may influence willingness to report certain behaviors.

While these results suggest that Hispanic men, especially if Mexican-born, may be less prone to abuse family pets, it is still important to consider the impact of family pets when providing services to families experiencing IPV. Even if an abusive partner has not actually harmed a pet, it does not mean he will not at some point in the future, particularly if his partner decides to leave. Moreover, the mere threat of harm to a pet is a form of abuse and is sufficient to cause distress to mothers and children and to potentially prevent a woman from seeking help. More research is needed to understand these lower rates of animal abuse, what factors may have contributed to the abuse, and whether those factors differ from those of non-Mexican or non-Hispanic abusers.

In summary, this study found that among families experiencing IPV, where the mother sought DV services, a significant proportion of the mothers’ partners threatened to harm or actually harmed a family pet. Rates of actual harm to pets were lower for Hispanic men, particularly if Mexican-born. Considering that the United States has a significant proportion of Hispanics and Mexican immigrants, it may be worthwhile to explore the topics of IPV and animal abuse within these populations.

Footnotes

Acknowledgements

We would like to thank the advocates who recruited and interviewed the survivors.

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) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: R01 HD066503.

Author Biographies

Christie A. Hartman is the data manager and statistician for the National Institute of Child Health and Human Development (NICHD) project “Children Exposed to Intimate Partner Violence and Animal Abuse” at the University of Denver. She received her PhD in psychology from the University of Colorado (CU), where she studied the genetic contributions to attention deficit/hyperactivity disorder (ADHD) and conduct disorder. She subsequently worked at CU as a scientist and professor studying substance dependence in adolescents and young adults.

Tina Hageman, MSW, MBA, is a clinical assistant professor at the University of Denver Graduate School of Social Work, where she teaches courses in organizational leadership. She has worked for more than 20 years in social services and nonprofit management, primarily in the fields of violence against women and crisis intervention. She has directed numerous programs including rape crisis services, domestic violence outreach counseling, battering intervention services, supervised visitation, and school-based prevention services.

James Herbert Williams, professor and Milton Morris Endowed Chair, is dean of the Graduate School of Social Work at University of Denver. His research and scholarship have focused on K-12 academic performance, youth violence, delinquency prevention, adolescent substance use, race and gender differences, and mental health service needs and utilization patterns in urban schools. He has also published widely in the area of health promotion, disease prevention among African American women, and, most recently, conflict and human security.

Frank R. Ascione is a scholar-in-residence at the University of Denver. His recent research examines the common roots of violence toward people and animals and is directed at identifying an early indicator of at-risk status in children. He has collaborated with human services, social work and child development staff working with abused children, with youth corrections personnel, and with state shelters for women who are battered. He has provided information or testimony for many state legislatures regarding cruelty to animals legislation, and has served on the American Psychological Association’s Presidential Task Force on Violence and the Family.

References

Ahrens

C. E.

Isas

Rios-Mandel

L. C.

Lopez

(2010). Talking about interpersonal violence: Cultural influences on Latinas’ identification and disclosure of sexual assault and intimate partner violence. Psychological Trauma: Theory, Research, Practice, and Policy, 2, 284-295.

Ascione

F. R.

(1997). Battered women’s reports of their partners’ and their children’s cruelty to animals. Journal of Emotional Abuse, 1, 119-133.

Ascione

F. R.

(2011). Pet Treatment Survey (PTS). Denver, CO: University of Denver.

Ascione

F. R.

Weber

C. V.

(1995). Battered Partner Shelter Survey (BPSS). Logan, UT: Utah State University.

Ascione

F. R.

Weber

C. V.

Thompson

T. M.

Heath

Maruyama

Hayashi

(2007). Battered pets and domestic violence: Animal abuse reported by women experiencing intimate violence and by non-abused women. Violence Against Women, 13, 354-373.

Becker

K. D.

Stuewig

Herrera

V. M.

McCloskey

L. A.

(2004). A study of firesetting and animal cruelty in children: Family influences and adolescent outcomes. Journal of the American Academy of Child & Adolescent Psychiatry, 43, 905-912.

Carlisle-Frank

Frank

J. M.

Nielsen

(2004). Selective battering of the family pet. Anthrozoös, 17, 26-42.

Cho

Velez-Ortiz

Parra-Cardona

J. R.

(2014). Prevalence of intimate partner violence and associated risk factors among Latinos/as: An exploratory study with three Latino subpopulations. Violence Against Women, 20, 1041-1058.

Currie

C. L.

(2006). Animal cruelty by children exposed to domestic violence. Child Abuse & Neglect, 30, 425-435.

10.

Dadds

M. R.

Whiting

Bunn

Fraser

J. A.

Charlson

J. H.

Pirola-Merlo

(2004). Measurement of cruelty in children: The Cruelty to Animals Inventory. Journal of Abnormal Child Psychology, 32, 321-334.

11.

DeGue

DiLillo

(2009). Is animal cruelty a “red flag” for family violence? Investigating co-occurring violence toward children, partners, and pets. Journal of Interpersonal Violence, 24, 1036-1056.

12.

Faver

C. A.

Cavazos

A. M.

(2007). Animal abuse and domestic violence: A view from the border. Journal of Emotional Abuse, 7, 59-81.

13.

Faver

C. A.

Strand

E. B.

(2003). To leave or to stay? Battered women’s concern for vulnerable pets. Journal of Interpersonal Violence, 18, 1367-1377.

14.

Flynn

C. P.

(2000). Woman’s best friend: Pet abuse and the role of companion animals in the lives of battered women. Violence Against Women, 6, 162-177.

15.

Flynn

C. P.

(2011). Examining the links between animal abuse and human violence. Crime Law Soc Change, 55, 453-468.

16.

Loring

M. T.

Bolden-Hines

T. A.

(2004). Pet abuse by batterers as a means of coercing battered women into committing illegal behavior. Journal of Emotional Abuse, 4, 27-37.

17.

Melson

G. F.

Schwarz

R. L.

Beck

A. M.

(1997). Importance of companion animals in children’s lives—Implications for veterinary practice. Journal of the American Veterinary Medical Association, 211, 1512-1518.

18.

Pew Research Center. (2006, March 7). Gauging family intimacy: Dogs edge cats (dads trail both). A social trends report of the Pew Research Center. Washington, DC: Author.

19.

Risley-Curtiss

Holley

L. C.

Wolf

(2006). The animal-human bond and ethnic diversity. Social Work, 51, 257-267.

20.

Siegel

J. M.

(1995). Pet ownership and the importance of pets among adolescents. Anthrozoos, 8, 217-223.

21.

Simmons

C. A.

Lehmann

(2007). Exploring the link between pet abuse and controlling behaviors in violent relationships. Journal of Interpersonal Violence, 22, 1211-1222.

22.

Straus

M. A.

Hamby

S. L.

Boney-McCoy

Sugarman

D. B.

(1996). The Revised Conflict Tactics Scales (CTS2): Development and preliminary psychometric data. Journal of Family Issues, 17, 283-316.

23.

Vaughn

M. G.

DeLisi

Beaver

K. M.

Perron

B. E.

Terrell

Howard

M. O.

(2009). Correlates of cruelty to animals in the United States: Results from the National Epidemiologic Survey on Alcohol and Related Conditions. Journal of Psychiatric Research, 43, 1213-1218.

24.

Volant

A. M.

Johnson

J. A.

Gullone

Coleman

G. J.

(2008). The relationship between domestic violence and animal abuse: An Australian study. Journal of Interpersonal Violence, 23, 1277-1295.

25.

Walton-Moss

B. J.

Manganello

Frye

Campbell

J. C.

(2005). Risk factors for Intimate partner violence and associated injury among urban women. Journal of Community Health, 30, 377-389.