J Fam Viol DOI 10.1007/s10896-016-9809-z
ORIGINAL ARTICLE
Intimate Partner Violence in Interracial Relationships Douglas A. Brownridge 1
# Springer Science+Business Media New York 2016
Abstract Research from the U.S. suggests that interracial relationships tend to have an elevated risk of intimate partner violence (IPV). The purposes of this study were to examine IPV in interracial relationships in Canada and explore the extent to which indicators derived from potential explanations account for the elevated risk of IPV in interracial relationships. Data were from a nationally representative sample of more than 19,000 Canadians, of whom 399 were in an interracial relationship and 9,969 were in a non-interracial relationship. Results showed that individuals in interracial relationships faced a significantly elevated risk of IPV victimization. Logistic regression analyses suggested that the presence of children and short union duration were particularly important in accounting for 56 % of the elevated odds of IPV in interracial relationships. Interracial relationships are a vulnerable population and the implications of the elevated risk of IPV in these unions need to be considered in IPV prevention efforts. Keywords Interracial . Intimate partner violence . Visible minority An increasing proportion of relationships in both the United States and Canada involve interracial couples. From a context in which anti-miscegenation laws forbade interracial relationships in the U.S. until 1967 (Mississippi, South Carolina, and Alabama did not repeal their state-level anti-miscegenation * Douglas A. Brownridge
[email protected] 1
Department of Community Health Sciences, College of Medicine, Faculty of Health Sciences, University of Manitoba, Winnipeg, MB, Canada R3T 2N2
laws until 1987, 1998, and 2000, respectively; Potter and Thomas 2012), rates of interracial coupling have increased five-fold since 1970 (Hattery and Smith 2009), comprising 8.4 % of all marriages in 2010 (Taylor et al. 2012). In Canada, 4.6 % of couples were living in interracial relationships in 2011. This represented an 18 % increase from 2006 (3.9 %), and a 77 % increase from 1991 (2.6 %) (Milan et al. 2010; Statistics Canada 2014). The trend toward more interracial coupling has been referred to as a ‘quiet revolution’ in which couples in these relationships have defied their cultural indoctrination in the name of love (Root 2001). It is true that attitudes toward interracial relationships have changed, with 77 % of Americans (Potter and Thomas 2012) and 92 % of Canadians approving of these relationships (Bibby 2007). However, attitudes toward these relationships are not always commensurate with individuals’ views on the appropriateness of these relationships for themselves or their loved ones (Bonilla-Silva and Forman 2000), with many persons in interracial relationships facing prejudice, discrimination and even violent hate crimes (Hattery and Smith 2009; Perry and Sutton 2006). Despite the trend toward increasing interracial coupling, and the unique circumstances presented by cultural views of these unions, there is a relative paucity of research on these relationships. Some research suggests that persons in interracial relationships are less satisfied with their relationship and experience more conflicts than non-interracial relationships (Potter and Thomas 2012). There are also indications in the literature that interracial couples in the U.S. have higher rates of IPV than non-interracial relationships. Among the many questions that must be answered for a better understanding of this phenomenon, we need to know whether an elevated risk of IPV in interracial relationships is a problem that is unique to the U.S. context. Moreover, to guide violence prevention efforts research must also articulate and examine
J Fam Viol
potential explanations. The purposes of the current study are to broaden the context of research by examining the risk for IPV in interracial relationships in Canada and to explore the extent to which indicators derived from potential explanations account for this phenomenon.
Indications of an Elevated Risk for IPV in Interracial Relationships Mercy and Saltzman’s (1989) study of lethal IPV provided an early indication that interracial relationships were at elevated risk of IPV. Among the observations in their study, these researchers reported that, from 1978 to 1981, interracial marriages had a homicide rate that was 7.7 times that of monoracial marriages. Subsequently, there have been six studies that have indicated the extent to which interracial relationships face an elevated risk of non-lethal IPV. Research on non-lethal IPV is recent, commencing with Kernsmith and Craun’s (2008) study of weapon use in domestic violence incidents. They studied 369 reports to the San Diego Sheriff’s Department and found that persons in interracial relationships comprised a disproportionate amount of their sample (19.5 %). Fusco (2010) also used police reports from a U.S. suburban county to study this phenomenon. She reported that interracial relationships were about twice as likely as monoracial relationships to involve mutual assault (interracial = 31.2 %; ethnic minority = 16.2 %; White = 13.4 %). While Fusco’s (2010) study used a larger sample than that of Kernsmith and Craun (2008), both studies are limited in their generalizability due to sampling one small geographic area. As well, they are both based on police report data that is likely to involve the most severe violence (Brownridge 2009; Straus 1990), and so do not speak to the more common, and typically less severe, violence that does not come to the attention of the police. The remaining four studies were all based on national probability samples of the U.S. and measured IPV using the Conflict Tactics Scales (Straus 1979; Straus et al. 1996). Hattery (2009) used data from the National Violence Against Women Survey (NVAWS) to compare violence against women in relationships that involved Black men and White women to those in which both partners were Black. She reported that the interracial relationships in her sub-sample had between two and four times the rate of violence compared to the Black intraracial relationships. Of course, limiting the analysis to a Black/White comparison excludes other visible minorities (i.e., individuals who are non-White/Caucasian) that are in interracial relationships. Using the same survey, Carbone-Lopez (2013) compared interracial relationships to monoracial White and monoracial minority relationships and found that both interracial and monoracial minority relationships had higher rates of overall and severe violence
than monoracial White relationships. Both the Hattery (2009) and Carbone-Lopez (2013) studies did not assess male victimization in interracial relationships. As well, the data were collected in the mid-1990s and so may not speak to the current risk of IPV. Using the National Couples Alcohol Survey, Chartier and Caetano (2012) compared interracial relationships between Whites, Blacks, and Hispanics to monoracial relationships in each of these three groups. They found that interracial couples had higher rates of both overall and severe IPV than monoracial couples. As with the NVAWS, these data were from 1995. The study also excluded non-Black/Hispanic minorities, so it cannot comment on the risk for these visible minorities. As well, the sample of interracial couples was small (n = 116), which limited the identification of predictors of IPV for these couples. Finally, Martin et al. (2013) compared interracial couples to monoracial Black and monoracial White couples using Wave IVof National Longitudinal Study of Adolescent Health (Add Health). They found that interracial and monoracial black couples had higher levels of mutual IPV than monoracial White couples. The data in this study were more recent than the other national probability studies discussed above, with Wave IVof Add Health having been collected in 2007/2008. However, the study was limited by its exclusion of all non-Black visible minorities, the exclusion of unidirectional IPV, and the restricted age range (18–24) of the sample. While limited, overall these studies indicate that interracial relationships tend to have a higher risk for violence than monoracial relationships. This problem appears to be evident for both lethal and non-lethal IPV, it is identified across official and self-report studies, it occurs for both females and males, and it is present throughout the U.S. This begs the question of why interracial relationships tend to have an elevated risk of IPV.
Potential Explanations for Elevated Rates of IPV in Interracial Relationships Rational explanations for an elevated risk of violence in interracial relationships need to be examined to provide an understanding of such results. A failure to do so opens the door for ‘knee-jerk’ explanations that are based in prejudiced and stereotypical views of minorities. Indeed, interracial relationships are commonly viewed as inherently dysfunctional, unnatural, and doomed (Perry and Sutton 2008). There is no theory specifically designed to explain why interracial relationships face an elevated risk of violence. What follows is a synthesis of speculations for this relationship in the literature. Stress The most commonly conjectured explanation for intimate partner violence in interracial relationships concerns a
J Fam Viol
number of potential sources of stress that may be unique to these unions. Racism and discrimination may be pernicious to the relationship in a number of ways. Persons in interracial relationships may face disapproval and ostracism from family, friends and the larger community (Carbone-Lopez 2013; Martin et al. 2013; Potter and Thomas 2012; Troy et al. 2006). Indeed, couples in mixed-race relationships are more likely than couples in monoracial relationships to hide their relationship due to fear of rejection (Fusco 2010). It has also been reported that the relative rarity of interracial relationships may lead them to be perceived by others as ‘exotic’, which, in turn, leads to Bintense negative curiosity^ from others (Hattery and Smith 2009, p. 71). Cleary, the perceptions and reactions of others to the relationship could cause stress for the couple. In addition to aspects of racism from outside the relationship, it has also been pointed out that racism within the couple may impact the likelihood of IPV in interracial relationships. Specifically, it has been noted that the minority partner may suspect, or make accusations of, racism and that racist statements may be made out of spite during arguments (CarboneLopez 2013). As well, in the context of an interracial relationship where one partner is White there may be a failure on the part of that individual to fully understand the unique circumstances faced by their visible minority partner. Consequently, she/he may minimize the impact of racism and discrimination and the members of the couple may not feel comfortable communicating with one another about racism and discrimination (Fusco 2010). The presence of children may pose some unique stressors in interracial relationships. In Canada, interracial couples are more likely, if only slightly so (58 % vs. 54 %), to have children (Milan et al. 2010) and research on IPV has shown that interracial couples are more likely than non-interracial couples to have children present at the scene (Fusco 2010). One way in which the presence of children may be particularly likely to increase stress and conflict in interracial couples’ relationships stems from potential cultural differences in values and expectations over child rearing practices (Troy et al. 2006). As well, given societal racism, there may be a concern by Whites in these relationships that their children will face discrimination, with the potential of not being accepted by either racial grouping (Potter and Thomas 2012). Indeed, an added challenge faced by couples with bi-racial children is fostering their children’s self-acceptance (Hegar and Greif 1994). Disparate views between members of the couple on these unique circumstances may lead to communication difficulties about interactions between the couple which, in turn, could cause stress and conflict (HohmannMarriott and Amato 2008). Given that interracial relationships may be more prone to stress, there are a number of potential risk factors for IPV that can be linked to stress and which should be examined. For example, in addition to the presence of children, alcohol abuse
may be linked to stress (Farrington 1986). Men in interracial relationships have been found to be more likely to engage in binge drinking and to have social problems due to alcohol abuse (Chartier and Caetano 2012). Other research, however, has found that members of interracial couples are less likely to use substances during an IPV incident (Fusco 2010). As will be discussed below, there are some differing indications about the socioeconomic status of persons in interracial relationships. Since unemployment is a stressful life circumstance, this variable should be examined in the current study. Finally, depression is associated with stress (e.g., Schafer et al. 1998) and IPV (e.g., Haj-Yahia 2000). Given the aforementioned potential stressors, it is possible that those in interracial relationships are more likely to be depressed and that this could impact their risk of IPV. Heterogamy Perspective It has been argued that those who cross racial boundaries in mate selection may also be more likely to cross other boundaries, and these differences may affect their relationship (Hohmann-Marriott and Amato 2008). Dissimilarity on a number of socially important characteristics, or heterogamy, may result in more disagreements not only between partners, but also between the couple and their friends and family (Chartier and Caetano 2012). The status incompatibilities that result from these dissimilarities can also create a power imbalance in the relationship (Carbone-Lopez 2013), and status incompatibilities are known to be associated with an increased risk of IPV (e.g., Macmillan and Gartner 1999). These differences may include incompatibilities in education, age, religion (Carbone-Lopez 2013; Chartier and Caetano 2012; Hohmann-Marriott and Amato 2008), as well as differences in expectations about love, sex, and childrearing (Carbone-Lopez 2013). Complex Relationship History Another challenge faced by some couples in interracial relationships concerns a tendency for individuals in these unions to possess relationship histories that are more complex than are those of individuals in noninterracial couples. As Hohmann-Marriott and Amato (2008) have identified, persons in interracial relationships are more likely to have had a previous marriage or cohabiting relationship. Remarriages are less stable than first marriages (Furstenberg and Cherlin 1991; Troy et al. 2006) and this instability could be, in part, linked to the presence of stepchildren (Hohmann-Marriott and Amato 2008). Remarriages, particularly those involving stepchildren, have been shown to face an elevated risk of violence against women (Brownridge 2004; Daly et al. 1993). Thus, a tendency toward greater relationship complexity may also play a role in the elevated risk of violence in interracial relationships. Demographic Characteristics of Relationships There are several demographic factors that may play a role in an
J Fam Viol
elevated risk of violence in interracial relationships. It has been suggested that lower overall levels of education among ethnic minorities may result in lower levels of education, on average, in interracial relationships (Hohmann-Marriott and Amato 2008). Some research has identified low levels of education as a risk factor for IPV (cf., Capaldi et al. 2012), and so this may help to explain a higher rate of violence in interracial relationships. However, it has also been suggested that a shift occurred at the end of the 20th Century from having less welleducated individuals in interracial relationships to having more highly educated individuals choosing these relationships (Fryer 2007). Having more highly educated individuals in these relationships may override prejudices (Root 2001) and help to counter some of the potential stressors discussed above. In Canada, interracial couples tend to be of higher socioeconomic status than non-interracial couples: they are more likely to involve university-educated partners, to be employed, and to have higher incomes (Milan et al. 2010). The only national-level study of violence in interracial relationships that examined socioeconomic status found that income and education did not predict IPV in these couples (Chartier and Caetano 2012). As people age and begin to more seriously consider dating partners to be potential marriage partners, they are less likely to date interracially (Potter and Thomas 2012). Indeed, interracial marriage is less socially accepted than interracial dating and so some interracial couples may choose cohabitation rather than marriage to avoid some of the stressors identified above that may be exacerbated by marriage (Potter and Thomas 2012). Population data have demonstrated this difference in both the United States and Canada. In the year 2000 in the U.S., 5.7 % of marriages involved interracial relationships compared to 10.2 % of cohabiting relationships (Hattery 2009). In Canada, almost one-quarter of interracial relationships in 2006 were living common-law, whereas less than one-fifth of non-interracial couples lived common-law (Milan 2003). The analysis of Martin et al. (2013) found that those in interracial relationships who were cohabiting had a higher risk of IPV than their counterparts who were married (AOR = 1.49; p < .001 vs. AOR = 1.05; p < .01). Consistent with younger daters being more likely to date interracially, younger people in general are more approving of, and more likely to engage in, interracial relationships (Carbone-Lopez 2013; Chartier and Caetano 2012). Indeed, in Canada interracial relationships are most common among those aged 25–34 (Statistics Canada 2014). Young age is a well-established risk factor for IPV (e.g., Capaldi et al. 2012), and so youth may play a role in the elevated risk of IPV in interracial relationships. Perhaps due to the propensity to be young and to choose cohabitation, which often ends in either marriage or dissolution (Brownridge 2008a; Lichter et al. 2006), interracial relationships tend to have a shorter
average duration than non- interracial relationships (Chartier and Caetano 2012). Short union duration is also associated with an increased risk of IPV (e.g., cf. Brownridge 2009).
Materials and Methods The Data Set The data employed in this study were from Cycle 23 of Statistics Canada’s General Social Survey (GSS), which was conducted in 2009. The GSS consisted of a random sample of 19,422 women and men who were age 15 years or older. Respondents completed in-depth telephone interviews concerning the nature and extent of their criminal victimization, including experiences of IPV. Because this study concerned IPV by current partners, the subsample of heterosexual women and men who were married or living common-law at the time of the survey consisted of 10,368 individuals who were either in an interracial (defined as couples comprised of a White/Caucasian and non-White/Caucasian; n = 399) or noninterracial (defined as couples in which both members of the couple were White/Caucasian or non-White/Caucasian; n = 9, 969) intimate relationship at the time of the study. It is important to note that Aboriginal Canadians are a unique vulnerable population, with a unique history that has impacted their experiences of IPV (e.g., cf. Brownridge 2008b). These individuals were excluded from the current study and will be analyzed separately in a future study that will consider the special circumstances surrounding Aboriginal Peoples’ experiences of IPV. In all analyses, the results were weighted and bootstrapped using STATA 13 with the person weight and bootstrap weights provided by Statistics Canada. Measurement Indicators of Stress The number of children referred to how many children (including birth, adopted and step-children) under age 15 were living in the respondent’s household. Respondent’s and partner’s employment were determined on the basis of whether the main activity in the 12 months prior to the interview was working at a paid job or business (employed) or looking for work, caring for children or housework (unemployed). Household income was derived from a series of questions asking respondents about their best estimate of their total household income. Heavy drinking by the respondent and their partner was measured with a question that asked how many times in the prior month that they had consumed five or more drinks on the same occasion. Depression referred to whether the respondent reported having used prescribed or over-the-counter medications to help them get out of depression in the month before the interview.
J Fam Viol
Indicators of the Heterogamy Perspective Education incompatibility was measured by calculating the ratio of the respondent’s years of education to the couple’s total years of education. For the analyses, education incompatibility was grouped into the following categories: the respondent had much less education (ratio < 0.46); the respondent had less education (ratio = 0.46 to 0.49); the respondent and their partner had the same years of education (ratio = 0.50); the respondent had more education (ratio = 0.51 to 0.54); and the respondent had much more education (ratio > 0.54). Age incompatibility was derived from the respondent’s report of their own and their partner’s age. The following categories were created: partner is six or more years older; partner is 1 to 5 years older; partner and respondent are the same age; partner is 1 to 5 years younger; and partner is six or more years younger.
respondent’s current marital or common-law partner within a specified time frame preceding the interview. Hence, if respondents reported having experienced any of the aforementioned forms of violence within the specified time frame preceding the interview they were coded as having experienced IPV. Severe IPV victimization included being: kicked, bit or hit with a fist; hit with something that could hurt; beaten; choked; threatened with or having a knife or gun used, and; sexually assaulted in the 5-year period prior to the interview. Due to concerns about respondent confidentiality, for the severe violence comparison Statistics Canada rounded cell counts to a base of 10 and truncated the decimal values for the proportions. Two reference periods were used in the current study: 1-year and 5-year time frames.
Indicators of Complex Relationship History Previous marriage/common-law relationship referred to whether the respondent had ever been in any other marriage or common- law relationship with a person other than their current partner. Living in a step-family was determined by whether the respondent’s family structure included living in a stepfamily (with or without a common child) or living in a family that did not have children from a previous relationship.
To examine the prevalence of IPVand investigate independent variables by relationship type, descriptive analyses were conducted using cross-tabulations with chi-square tests of significance. Logistic regression was used for the multivariate analyses because it is an appropriate technique for predicting a dichotomous dependent variable from a set of independent variables. For model parsimony, only significant variables from the descriptive analyses were included in the multivariate analyses. Two sets of logistic regressions were conducted. The first involved direct logistic regressions with all of the predictor variables. These analyses allowed examination of the operation of each independent variable in the prediction of violent victimization for respondents in interracial and non-interracial relationships holding constant the effects of all other variables in the model. Also, these analyses, combined with t-tests to examine differences in odds ratios, allowed comparisons of the operation of these variables for respondents across interracial and noninterracial relationships. The second set of multivariate analyses were sequential logistic regressions. In these regressions, the first model contained only the interracial/non-interracial status variable without any controls. In the second model, this variable was entered along with variables derived from the stress explanation. In the third model, the interracial/non-interracial status variable was entered along with the heterogamy variables. In the fourth model, the interracial/non-interracial status variable was entered along with the demographic characteristics. Finally, in the fifth model, the interracial/non-interracial status variable was entered along with all of the predictor variables. These sequential logistic regressions allowed an examination of the extent to which controlling for the proxies of stress, heterogamy, and demographic characteristics accounted for the elevated odds of violence against respondents in interracial relative to non-interracial relationships.
Indicators of Demographic Characteristics Marital status referred to whether the respondent was living in a common-law relationship or married at the time of the survey. Age of the respondent was measured by asking the respondent their birth date and confirming their correct age. Partner’s age was derived from a question that asked respondents about the age of each member of their household. Duration of relationship was measured with a variable derived from the respondent’s report of how long they had been living with their marital or common-law partner. Respondent’s and partner’s education were measured by asking the respondent about the highest level of education that they and their partner had attained. Violence Ten behavioral items from a modified version of the Conflict Tactics Scales (CTS; Straus 1979) were used to measure violence. IPV was defined as acts of physical assault (having something thrown at you that could hurt; being pushed, grabbed, or shoved in a way that could hurt; being slapped; being hit with something that could hurt; being kicked, bit, or hit with a fist; being beaten; being choked; being threatened with or having a knife or gun used against the respondent), physical threat (being threatened to be hit with a fist or anything else that could hurt), and sexual assault (being forced into any sexual activity by being threatened, held down, or hurt in some way) perpetrated by the
Method of Analysis
J Fam Viol
Results Descriptive Analysis Violence by Relationship Type Results showed that respondents in an interracial relationship had the highest prevalence of IPV (7.8 %), followed by those in White monoracial relationships (3.7 %), and those in relationships in which both the respondent and their partner were a visible minority (3.0 %; p < 0.001). A comparison of the latter two groups showed that their respective prevalence rates of IPV were not significantly different (χ2 = 1.5, df = 2). Because the White monoracial and visible minority relationships did not differ in their prevalence of violence, they were combined in the remaining analyses and referred to as non-interracial relationships. Not only did interracial relationships have a higher rate of IPV in the 5-year time frame (7.8 % vs. 3.6 %; p < .001), but the elevated risk of IPV for interracial relationships was also evident in the year prior to the survey (4.6 % vs. 1.1 %; p < 0.001). Moreover, the elevated risk of IPV in interracial relationships held when examining only severe IPV victimization (3.3 % vs. 1.2 %; p < 0.001). Independent Variables by Relationship Type Table 1 contains the results of the cross-tabulations of the independent variables by interracial/non-interracial status. Only one of the variables derived from the stress explanation was significantly different for interracial and non- interracial relationships. Respondents in interracial relationships were much more likely to have one or two young children (p < .001). Both the education and age heterogamy variables showed significant differences between the two groups. Respondents in interracial relationships were more likely to have much less education than their partner (p < .001). Respondents in interracial relationships were also more likely than those in noninterracial relationships to have a partner that is six or more years younger or older, and they were less likely to have a partner of the same age (p < .001). Neither of the variables derived from the complex relationship history explanation showed significant differences between the two groups. In terms of demographic characteristics, all but the marital status variable showed significant differences between the two groups (p < .001). Those in interracial relationships tended to be younger, to have been in their relationship for less time, and to have higher levels of education.
children was associated with increased odds of IPV for respondents in interracial relationships, but not those in noninterracial relationships. This was evident regardless of whether the respondent had one (AOR = 3.930; p < .001), two (AOR = 3.308; p < .001) or three or more (AOR = 3.951; p < .001) children. Education incompatibility was, overall, not a strong predictor of IPV for those in interracial relationships, with the exception of the respondent having much more education than their partner being a protective factor (AOR = 0.242; p < .001). On the other hand, age incompatibility was a stronger predictor of IPV for those in interracial relationships. While having a partner that was six or more years older was associated with a similar positive increase in odds of violence for both groups, those in interracial relationships with a partner that was 1-to-5 years older (AOR = 2.812; p < .001), 1-to-5 years younger (AOR = 2.757; p < .001), or six or more years younger (AOR = 4.007; p < .001) each faced increased odds of reporting having experienced IPV. Moreover, the t-tests showed that the latter three effects were stronger for those in interracial than non-interracial relationships. Finally, with respect to demographic characteristics, the only significant variable for those in interracial relationships was union duration. For each additional year that a couple had been together, the odds of IPV were reduced by 6 % for those in interracial relationships (p < .001). Sequential Logistic Regressions Table 3 contains the results of sequential logistic regressions on the 5-year prevalence of violence. As shown in Table 3, the difference in prevalence between respondents in interracial and non-interracial relationships translated to those in interracial relationships having 124 % greater odds of IPV compared to those in noninterracial relationships. Controlling for stress indicators reduced the elevated odds for respondents in interracial relationships by 21 % (t = 1.96; p = .05). There was no reduction in odds when controlling for the heterogamy variables (t = −1.47). The demographic characteristics reduced the elevated odds by 53 % (t = 5.40; p < .001). When indicators of all of the explanations were entered, the reduction in the elevated odds for respondents in interracial relationships was greatest (56 %; t = 5.69; p < .001). However, even after controlling for all of these variables, respondents in interracial relationships had 69 % greater odds of violence than those in non- interracial relationships (p < .001).
Multivariate Analysis
Discussion
Direct Logistic Regressions for Interracial and Noninterracial Relationships Table 2 contains the results of the direct logistic regressions for those in interracial and noninterracial relationships on the 5-year prevalence of violence. Controlling for all other variables in the models, having young
Consistent with past studies that have compared interracial to various combinations of non- interracial relationships, respondents in interracial relationships in the current study faced an elevated risk of IPV. Indeed, those in interracial relationships had twice the rate of IPV victimization in the 5 years prior to
J Fam Viol Table 1 Independent variables by interracial/non-interracial status (%)
Independent variables
Interracial
Non-interracial
Stress Respondent’s employment Did not work in past year
13.6
14.8
86.4
85.2
18.2
15.5
81.8
84.5
0–29,999
5.0
7.7
30,00–49,999 50,000–79,999
12.4 26.7
14.6 25.5
80,000 or more
56.0
52.2
Worked past year Partner’s employment Did not work in past year Worked past year Respondent’s household income
Respondent’s heavy drinking (past month) Yes
29.5
33.5
No Partner’s heavy drinking (past month)
70.5
66.5
Yes No Children < 15 None
16.3 83.7
18.8 81.2
48.7
65.4
23.1 23.3 4.9
15.0 14.7 5.0***
4.4 95.6
5.9 94.1
17.7 15.0 35.9 13.1 18.3
13.8 12.8 36.7 19.4 17.3***
16.7 33.5 9.6 26.8 13.4
10.5 32.2 12.3 33.8 11.2***
8.4 91.6
7.1 92.9
11.8 88.2
10.5 89.5
16.4
17.6
One Two Three or more Depression Yes No Heterogamy perspective Education compatability Respondent has much less education Respondent has less education Respondent and partner same education Respondent has more education Respondent has much more education Age compatability Partner is 6+ years older Partner is 1–6 years older Partner and respondent same age Partner is 1–5 years younger Partner is 6 or more years younger Complex relationship history Previous marriage/common-law union Yes No Stepfamily Yes No Demographic characteristics Marital status Common-law
J Fam Viol Table 1 (continued) Independent variables Married
Interracial
Non-interracial
83.6
82.4
Respondent’s age 15–34
29.3
18.8
35–54 55 and older
56.4 14.4
45.9 35.3***
15–34 35–54
25.1 59.2
17.8 47.5
55 and older
15.7
34.7***
20.1 32.2
10.3 18.6
47.7
70.8***
8.1 5.8
13.2 15.0
Partner’s age
Duration of relationship Less than 4 years 4–9 years 10 or more years Respondent’s education Less than high school High school
14.6
11.7
Community college diploma/certificate University degree Partner’s education Less than high school High school Some post-secondary
Some post-secondary
27.0 44.6
30.5 29.5***
3.5 19.3 10.6
14.2 25.9 9.1
Community college diploma/certificate University degree
19.0 47.7
20.3 30.4***
* p < 0.05; ** p < 0.01; *** p < 0.001 (p values refer to χ2 tests of significance)
the survey and four times the rate in the year prior to the survey. This pattern also held when examining severe IPV victimization. As shown in the review of the literature, an elevated risk for IPV in interracial relationships has been identified in the U.S. The current study of IPV in interracial relationships is the first to be conducted outside of the U.S and has demonstrated that an elevated risk of IPV in interracial relationships is not limited to the U.S. context. Although no theory has been developed for understanding this phenomenon, the current study organized speculations for this relationship into four categories: stress; heterogamy perspective; complex relationship history; and demographic characteristics of relationships. Proxy indicators derived from these explanations that were available in the 2009 GSS were identified and compared across interracial and non-interracial relationships. Results showed that there were significant differences on several of these proxy indicators, with the exception of those derived from complex relationship history. Those in interracial relationships were more likely to have one or two children and having any number of children was associated with increased odds of IPV victimization for respondents in these unions. By contrast, for non-interracial
relationships, having children tended to be associated with reduced odds of IPV victimization. Controlling for this variable reduced the elevated odds of IPV victimization by 21 %. Clearly, based on these data, it would appear that there is something about having children in interracial relationships that is associated with their increased likelihood of IPV. The review of the literature suggests that this could be a consequence of cultural differences in child rearing practices that lead to stress and conflict, and with concerns about these children being accepted and discriminated against. However, it is important at this juncture to emphasize that the presence of children is not a measure of stress and so caution is needed when extrapolating from these results. With respect to the heterogamy perspective, interracial relationships were not particularly educationally incompatible relative to non-interracial relationships and this variable did not predict increased odds of IPV for those in interracial relationships. Interracial relationships did tend to be more age incompatible, and age incompatibility was associated with particularly increased odds of violence for respondents in interracial relationships. However, controlling for these incompatibility variables did not reduce the
J Fam Viol Table 2 Results of logistic regressions on 5-year prevalence of violence for interracial and non-interracial relationships
Covariates
Interracial Odds ratio
Non-interracial Odds ratio
t
None One
1.000 3.930***
1.000 0.888**
15.835***
Two
3.308***
0.957
13.609***
3.951***
0.824**
11.617***
0.714
3.028***
−6.213***
Stress Children < 15
Three or more Heterogamy perspective Education incompatibility Respondent has much less education Respondent has less education
1.283
1.788***
−2.422*
Respondent and partner same education Respondent has more education
1.000 0.866
1.000 0.883**
−0.150
Respondent has much more education
0.242***
0.408***
−2.609**
1.329
1.665***
−1.598
2.812*** 1.000 2.757***
1.018 1.000 1.123*
8.402***
4.007***
1.007
9.3241***
1.017 0.941*** 0.915
0.947*** 0.997 1.336***
9.953*** −8.764*** −5.590***
0.967 0.086** 1053***
0.778*** 0.211*** 3274***
4.122***
Age incompatibility Partner is 6+ years older Partner is 1–5 years older Partner and respondent same age Partner is 1–5 years younger Partner is 6 or more years younger Demographic characteristrics Partner’s age Duration of relationship Respondent’s education Partner’s education Constant χ2
6.720***
*p < 0.05; **p < 0.01; ***p < 0.001
elevated odds of IPV for those in interracial unions. There are several potential incompatibilities for which there were no proxy measures available in the data used in the current study, including incompatibilities over expectations about love, sex and childrearing. Such incompatibilities should be examined in future research. Given the finding that the Table 3
presence of children is associated with an elevated risk of IPV in these unions, an examination of incompatibilities over child rearing would appear to be a particularly fruitful avenue for further investigation. Finally, in terms of demographic characteristics of relationships, those in interracial relationships tended to be
Results of sequential logistic regressions on 5-year prevalence of violence
Covariates
Interracial/ non-interracial Odds ratio
Interracial/ non-interracial Interracial 2.243*** Non-interracial 1.000
Stress Odds ratio
Heterogamy Odds ratio
Demographic characteristics Odds ratio
Full model Odds ratio
2.037*** 1.000 BLOCK1
2.411*** 1.000
1.712*** 1.000
1.688*** 1.000 BLOCK1 BLOCK2 BLOCK3 0.155*** 3722***
BLOCK2 Constant χ2 ***p < 0.001
0.377*** 272***
0.286*** 1485***
0.035*** 500***
BLOCK3 0.304*** 2880***
J Fam Viol
young, to have been together for less time, and to be more highly educated. Controlling for all other variables in the model, young age did not predict IPV for interracial relationships, nor did education. But union duration was a significant predictor of IPV in these unions. For each additional year that an interracial couple had been together, the respondent’s odds of reporting having experienced IPV were reduced by 6 % (p < .001). Controlling for these demographic characteristics produced the greatest reduction in the elevated odds of IPV for interracial relationships (53 %). Thus, it would appear that the tendency for interracial couples to have been together for a shorter period of time accounted for a large portion of the elevated risk of IPV in these unions. Perhaps the tendency to have shorter duration relationships results in persons in these unions, on average, to have had less time to establish more appropriate ways of dealing with conflicts. Thus, combined with the potentially greater stressors they face, this places them at increased risk for IPV. Indeed, the results showed that the greatest reduction in the elevated odds of IPV resulted when both union duration and the presence of children variables were included in the model. Although the variables in the current study accounted for 56 % of the elevated odds of IPV for interracial relative to non-interracial relationships, after adjusting for these variables respondents in interracial unions still had 69 % elevated odds of reporting having experienced IPV. It is noteworthy that the proxies which were available in the data accounted for a substantively significant proportion of the elevated odds of IPV, particularly in light of that fact that these variables were less than ideal indicators derived from explanations identified in the literature to account for this relationship. Nevertheless, as they relate to speculations in the literature for an elevated risk of IPV in interracial relationships, due to measurement limitations the findings of the current study must be regarded as preliminary in nature. Indeed, the variables that were available in the data do not serve as formal measures of constructs derived from speculations for this relationship. What is clear from this study is that future research is needed that is specifically designed to examine the elevated risk of IPV in interracial relationships. Validated measures of relevant concepts and constructs, such as stress, couple communication, differences in expectations, and racism and discrimination, need to be examined. As well, there is no theory specifically developed to understand the phenomenon of an elevated risk of IPV in interracial relationships, and such a theory would be helpful for guiding future research. In addition to being limited by proxy measures available in the data, there were other noteworthy limitations in this study. The data were collected by telephone survey and so potential respondents that did not have a phone were excluded. The survey was also limited to respondents that could speak in English or French. As well, respondents were asked only about victimization, and so it was not possible with the data
to establish the extent to which IPV was bi-directional using these data. Finally, due to sample size restrictions it was not possible to disaggregate interracial relationships by different ethnic combinations. Despite these limitations, the current study contributed new knowledge to the literature. It has demonstrated that the elevated risk for IPV in interracial relationships found in the U.S. context also exists in the Canadian context. Unlike past studies, which have tended to focus on either Black/White or Black/White/Hispanic combinations, the analyses in the current study included non-Black and non-Hispanic visible minorities. This is particularly relevant because in the U.S., Whites who marry exogamously are most likely to marry someone of Asian descent (Hohmann-Marriott and Amato 2008) and in Canada those who are Japanese are the most likely ethnic minority to intermarry (Milan 2003). The current study also included data on both female and male IPV victimization, and so provided a more complete picture of IPV in interracial relationships than previous studies that have focused only on female victims. This was also the first study to consolidate speculations for the elevated risk of IPV in interracial unions, and to attempt to examine proxy indicators derived from each of these potential explanatory factors. As well, the current study used data that were more recent than previous studies, most of which relied on data from the mid-1990s. While attitudes toward interracial relationships have changed, the current study demonstrated that these relationships continue to face an elevated risk of IPV. It bears pointing out that the data also showed that most respondents in interracial relationships, 92 %, did not report having experienced IPV in the 5 years prior the study. Thus, the vast majority of these individuals did not experience IPV. Nevertheless, those in interracial relationships were at least twice as likely as their counterparts in non-interracial relationships to be victimized by their partner, suggesting that more work is needed to address the unique circumstances faced by this vulnerable population. In addition to further research designed to understand why individuals in interracial relationships are more likely to experience IPV, stakeholders need to be aware of this phenomenon and to consider its implications for violence prevention.
Acknowledgments This research was supported by funds to the Canadian Research Data Centre Network (CRDCN) from the Social Science and Humanities Research Council (SSHRC), the Canadian Institute for Health Research (CIHR), the Canadian Foundation for Innovation (CFI) and Statistics Canada. Although the research and analysis are based on data from Statistics Canada, the opinions expressed do not represent the views of Statistics Canada or the Canadian Research Data Centre Network (CRDCN).
J Fam Viol
References Bibby, R. W. (2007). Racial Intermarriage: Canada and the U.S. [Press release]. Retrieved from http://www.reginaldbibby.com/images/PC_ 9_RACIAL_INTERMARRIAGE_AUG2907.pdf. Bonilla-Silva, E., & Forman, T. A. (2000). BI am not a racist but…^: mapping white college students’ racial ideology in the USA. Discourse and Society, 11(1), 50–85. doi:10.1177/0957926500011001003. Brownridge, D. A. (2004). Male partner violence against women in stepfamilies: an analysis of risk and explanations in the Canadian milieu. Violence and Victims, 19(1), 17–36. Brownridge, D. A. (2008a). The elevated risk for violence against cohabiting women: a comparison of three nationally representative surveys of Canada. Violence Against Women, 14(7), 809–832. doi:10. 1177/1077801208320368. Brownridge, D. A. (2008b). Understanding the elevated risk of partner violence against aboriginal women: a comparison of two nationally representative surveys of Canada. Journal of Family Violence, 23(5), 353–367. doi:10.1007/s10896-008-9160-0. Brownridge, D. A. (2009). Violence against women: vulnerable populations. New York: Routledge. Capaldi, D. M., Knoble, N. B., Shortt, J. W., & Kim, H. K. (2012). A systematic review of risk factors for intimate partner violence. Partner Abuse, 3(2), 231–280. doi:10.1891/1946-6560.3.2.231. Carbone-Lopez, K. (2013). Across racial/ethnic boundaries: investigating intimate violence within a national sample. Journal of Interpersonal Violence, 28(1), 3–24. doi:10.1177/0886260512448850. Chartier, K. G., & Caetano, R. (2012). Intimate partner violence and alcohol problems in interethnic and intraethnic couples. Journal of Interpersonal Violence, 27(9), 1780–1801. doi:10.1177/ 0886260511430392. Daly, M., Singh, L., & Wilson, M. (1993). Children fathered by previous partners: a risk factor for violence against women. Canadian Journal of Public Health, 84(3), 209–210. Farrington, K. (1986). The application of stress theory to the study of family violence: principles, problems, and prospects. Journal of Family Violence, 1(2), 131–132. Fryer, R. G., Jr. (2007). Guess who’s been coming to dinner? Trends in interracial marriage over the 20th century. Journal of Economic Perspectives, 21(2), 71–90. Furstenberg, F. F., Jr., & Cherlin, A. J. (1991). Divided families: what happens to children when parents part. Cambridge: Harvard University Press. Fusco, R. A. (2010). Intimate partner violence in interracial couples: a comparison to white and ethnic minority monoracial couples. Journal of Interpersonal Violence, 25(10), 1785–1800. doi:10. 1177/0886260509354510. Haj-Yahia, M. M. (2000). Implications of wife abuse and battering for self-esteem, depression, and anxiety as revealed by the second Palestinian national survey on violence against women. Journal of Family Issues, 21(4), 435–463. doi:10.1177/019251300021004002. Hattery, A. J. (2009). Intimate partner violence. Lanham: Rowman & Littlefield. Hattery, A. J., & Smith, E. (2009). Race and intimate partner violence: Violence in interracial and intraracial relationships. In E. Smith & A. J. Hattery (Eds.), Interracial relationships in the 21st century (pp. 67–88). Durham: Carolina Academic Press. Hegar, R. L., & Greif, G. L. (1994). Parental abduction of children from interracial and cross- cultural marriages. Journal of Comparative Family Studies, 25(1), 135–142. Hohmann-Marriott, B. E., & Amato, P. (2008). Relationship quality in interethnic marriages and cohabitations. Social Forces, 87(2), 825– 855. doi:10.1353/sof.0.0151.
Kernsmith, P., & Craun, S. W. (2008). Predictors of weapon use in domestic violence indicents reported to law enforcement. Journal of Family Violence, 23(7), 589–596. doi:10.1007/ s10896-008-9181-8. Lichter, D. T., Qian, Z., & Melott, L. M. (2006). Marriage or dissolution? Union transitions among poor cohabiting women. Demography, 43(2), 223–240. doi:10.1353/dem.2006.0016. Macmillan, R., & Gartner, R. (1999). When she brings home the bacon: labor-force participation and the risk of spousal violence against women. Journal of Marriage and the Family, 61(4), 947–958. doi: 10.2307/354015. Martin, B. A., Cui, M., Ueno, K., & Fincham, F. D. (2013). Intimate partner violence in interracial and monoracial couples. Family Relations, 62(1), 202–211. doi:10.1111/j.1741-3729.2012.00747.x. Mercy, J. A., & Saltzman, L. E. (1989). Fatal violence among spouses in the United States, 1976–85. American Journal of Public Health, 79(5), 595–599. Milan, A. (2003). Would you live common-law? Canadian social trends (pp. 2–6). Ottawa: Statistics Canada. Milan, A., Maheux, H., & Chui, T. (2010). A portrait of couples in mixed unions Canadian social trends (pp. 70–80). Ottawa: Statistics Canada. Perry, B., & Sutton, M. (2006). Seeing red over black and white: popular and media representations of inter-racial relationships as precursors to racial violence. Canadian Journal of Criminology and Criminal Justice, 48(6), 887–904. Perry, B., & Sutton, M. (2008). Policing the colour line violence against those in intimate interracial relationships. Race, Gender and Class, 15(3–4), 240–261. Potter, H., & Thomas, D. T. (2012). BWe told you that’s how they are^: responses to white women in abusive intimate relationships with men of color. Deviant Behavior, 33(6), 469–491. doi:10.1080/ 01639625.2011.636661. Root, M. P. P. (2001). Love’s revolution: interracial marriage. Philadephia: Temple University Press. Schafer, R. B., Wickrama, K. A. S., & Keith, P. M. (1998). Stress in marital interaction and change in depression: a longitudinal analysis. Journal of Family Issues, 19(5), 578–594. doi:10.1177/ 019251398019005005. Statistics Canada. (2014). Mixed unions in Canada: national household survey (NHS), 2011. Ottawa: Statistics Canada. Straus, M. A. (1979). Measuring intrafamily conflict and violence: the conflict tactics (CT) scales. Journal of Marriage and the Family, 41(1), 75–88. Straus, M. A. (1990). Injury and frequency of assault and the Brepresentative sample fallacy^ in measuring wife beating and child abuse. In M. A. Straus & R. J. Gelles (Eds.), Physical violence in American families: risk factors and adaptations to violence in 8,145 families (pp. 75–91). New Brunswick: Transaction. Straus, M. A., Hamby, S. L., Boney-McCoy, S., & Sugarman, D. B. (1996). The revised conflict tactics scales (CTS2): development and preliminary psychometric data. Journal of Family Issues, 17(3), 283–316. Taylor, P., Wang, W., Parker, K., Passel, J. S., Patten, E., & Motel, S. (2012). The rise of intermarriage: rates, characteristics vary by race and gender social & demographic trends. Washington, D.C.: Pew Research Center. Troy, A. B., Lewis-Smith, J., & Laurenceau, J.-P. (2006). Interracial and intraracial romantic relationships: the search for differences in satisfaction, conflict, and attachment style. Journal of Social and P e r s o n a l R e l a t i o n s h i p s , 2 3 ( 1 ) , 6 5 – 8 0 . d o i : 1 0 . 11 7 7 / 0265407506060178.