Journal of Family Violence

, Volume 23, Issue 1, pp 25–35

American Indian and European American Women’s Perceptions of Domestic Violence


    • University of Nebraska
    • University of Arizona
  • Cynthia Willis Esqueda
    • University of Nebraska

DOI: 10.1007/s10896-007-9126-7

Cite this article as:
Tehee, M. & Esqueda, C.W. J Fam Viol (2008) 23: 25. doi:10.1007/s10896-007-9126-7


American Indian and European American women’s definitions and perceived causes for domestic violence were examined. Attitudes towards violence and battering as it relates to the self were measured with two scales. As predicted, results indicated American Indian women and European American women held different conceptualizations of what constitutes domestic violence and different notions concerning the cause of domestic violence. Also, American Indian women were more attuned to external causes for violence, while European American women referred to internal explanations for such violence. Differences in social and psychological histories of violence and attitudinal orientations toward violence were indicated. Legal and health system changes are recommended in order to combat violence in Indian country.


Domestic violenceAmerican Indian womenEuropean American womenAmerican Indian culture

There are over 2 million American Indians1 and 562 federally recognized tribes in the USA (United States Bureau of Indian Affairs; United States Census Bureau 2005). Tribes possess unique histories, traditions, and cultures, and they deal with a myriad of issues (Boczkiewicz 2003; Carpenter 1999; Cole 2005; Eig 2002; Overend 2004; Selden 2002).

Violence is one commonality for Indian people and tribes (Perry 2004; Wallace et al. 1997). American Indians experience more interpersonal violence than any other US ethnic group (Greenfield and Smith 1999; Perry 2004). And Indian women are at high risk for domestic violence. As an indication of the prevalence of interpersonal violence, “Indian love” has become slang for such abuse.

Research indicates American Indian women experience domestic violence at high levels (Bohn 2003; Chester et al. 1994; Fairchild et al. 1998; Hamby 2000; Halinka Malcoe and Duran 2004; Halinka Malcoe et al. 2004; Murphy et al. 2003). Further, American Indian women have reported higher rates of intimate partner violence victimization compared to women from other ethnic groups (Tjaden and Thoennes 1998a, b). From the extant literature, Indian women are at higher risk for domestic violence than non-Indian women (Oetzel and Duran 2004), and Native American women experience violent victimization at twice the rate of African American women (Dugan and Apel 2002).

Compared to other ethnicities, Indian women experience physical abuse more often than verbal abuse. In addition, Indian women sought medical help more frequently for abuse than other ethnicities (Krishnan et al. 1997). Domestic violence leads to severe injuries requiring medical attention for Indian women (Bohn 1998). A high percentage of American Indian women have a history of abuse, and their partners have a history of abuse as well (Bohn 2003; Krishnan et al. 1997). As would be expected, then, interpersonal violence is a common occurrence in American Indians’ lives, and is increasing both on and off reservations (Allen 1985).

Although current rates are high, domestic violence is a recent phenomenon in American Indian communities. Traditionally, American Indian communities did not accept such violence. One common theme is that “it is the mothers, not the warriors, who create a people and guide their destiny” (Standing Bear 1978, dedication). Although gender roles varied among tribes, many were matrilocal and matrilineal. Regardless of tribe, American Indian culture provided women with the inherent opportunity to have social and familial power. The importance of Indian women’s economic, spiritual, political, and social power has been noted (Chester et al. 1994). In general, tribes were more egalitarian than European colonial communities (Allen 1986; Plane 2000; Shoemaker 1995).

American Indian women’s inherent powers and family/clan connections made a community response to domestic violence inevitable. Community responses to such violence were immediate and definitive. In many communities divorce was easily accomplished when family violence occurred (Murray 1998; Perdue 1998), and the community could take away the abuser’s honored roles, and ostracize or even enforce exile of an abuser. In a psychological system that demanded (and demands) connections to spirituality, imparted by knowledge learned from family elders and sacred tribal members and entities, exile was a punishment that was worse than death. Thus, domestic violence was thwarted or severely curtailed through actions by the victim or by the community.

Legal and social policies of the USA towards American Indian tribes and their members have impacted the traditional systems of community and family life. The introduction of European legal and education systems and the elimination of traditional peace keeping practices allowed domestic violence to thrive. The adoption of western European criminal justice systems, lifestyles, and values left women without the inherent powers that made domestic violence a rare occurrence (Allen 1986).

After European contact, Indian people were characterized as savage, inferior, and uncivilized (Bohn 1998). American Indian women were not overlooked in the dominant culture’s biased assessments. The two icons of drudge or slave and Indian princess have been noted at least since the seventeenth century (Anderson 1996; Berger 1997; Riley 1984). These stereotypes have psychological consequences for women, treatment of women, and violence towards them, and such stereotypes have been labeled the “prostitute-princess syndrome” (Medicine 1988). It is possible American Indian women have incorporated the dominant culture’s beliefs that they are of less value (Mihesuah 1996), and stronger than European American women (Allen 1985). Therefore, actions that European American women view as abuse may not be considered so by American Indian women, and they may have self conceptualizations that fit battered women.

Although rates of domestic violence are high in the Indian community, research and treatment programs have failed to consider Indian definitions of, and responses to, such violence (Green 1980). Chester et al. (1994) stated that “definitions of behaviors for denoting abuse of women by male partners have also been problematic. Various forms of physical and verbal abuse have been incorporated into terms, such as conjugal violence, domestic violence, and family violence,” (p. 250). Although research on domestic violence development, expression, and control has proliferated, most of this research has concentrated on a European American perspective (Kasturirangan et al. 2004). Without asking for Native women’s perceptions of, and definitions for, domestic violence, research directed at theory generation, research implementation, and social policy promotion will remain stilted. Consequently, the purpose of this project was to investigate what constitutes domestic violence according to American Indian and European American women. In other words, how do American Indian and European American women view domestic violence, and what actions are considered to identify domestic violence?

Based on unique historical and cultural experiences, we hypothesized American Indian women would focus on behavioral and contextual aspects of violence, while European American women would focus on dispositional aspects of domestic violence. American Indian women would also relate that such violence was not part of cultural tradition, but current estimates would reflect high community incidence rates. We anticipated European American women would relate that domestic violence has been part of their culture, but their current estimates of community incidence rates would be lower than those for American Indian women.

Further, we anticipated that perceptions of domestic abuse between American Indian women and European American women would produce different notions of causation. A foundational principal for assigning causation is the use of internal, dispositional and external, situational references (Heider 1958, 1984, personal communication). Consequently, we made predictions based on attribution styles, whereby those from western, individualistic cultures tend to use dispositional attributions concerning behavior, while those from communal cultures tend to use more situational approaches (Triandis 2001; Zárate et al. 2001). Traditionally, xAmerican Indians have relied on environmental and communal or social contexts to understand behavior, while European Americans have relied on internally driven, dispositional traits to explain behavior. We predicted American Indian women would focus on actions and external/contextual factors for domestic abuse definitions and explanations, whereas European American women would focus on internal issues and dispositions.

Differences in conceptualizations of domestic violence among American Indian and European American women may stem from domestic violence attitude differences, as well. We examined attitudinal differences using two domestic violence attitudinal measures. The Battered Women’s Scale (BWS) measures self concept and gender role trait attributions as they pertain to battering, “showing that the experience of being a survivor of a tyrannical battering situation changes at least part of a woman’s self-concept” (Schwartz and Mattley 1993, p. 286). Thus, those who score high on the scale conceptualize themselves to be lower in agency and are inclined to believe they are easily dominated and less forceful. Two outcomes are possible with regard to how American Indian women view themselves, in comparison to European American women. Based on historical and cultural experiences, American Indian women may have come to internalize the devaluation of the larger US culture and score high on the BWS. In contrast, American Indian women may conceive of themselves as forceful and agentic, based on their traditional valued roles in American Indian culture, and score lower on the BWS in comparison to European American women.

Originally, the Attitudes Towards Violence Scale or ATVS (Funk et al. 1999) was designed to examine willingness to use and acceptance of violence for adolescents. This scale was found to have two subscales, Culture of Violence and Reactive Violence.

Culture of Violence items reflect the conviction that the world is a dangerous place, where the best way to ensure survival is to be vigilant and ever prepared to take the offensive. Individuals endorsing these items view themselves as members of a local culture of violence and thus as both a likely victim and perpetrator of violence (Funk et al. 1999, p. 1129).

Reactive Violence items reflect those attitudes which are related to an individual’s response to an immediate threat. This subscale includes items that indicate the following: having violent behaviors in one’s repertoire, being willing to act in a violent manner, and endorsing the actual choice of a violent response (Funk et al. 1999, p. 1129).

An examination of the items indicated the scale would be useful in explicating differences between our two samples and their approach to violence in their lives. Based on differences in historical experiences and current prevalence, we anticipated that American Indian women would indicate higher violence acceptance on both the cultural violence and reactive violence scales, compared to European American women.

Thus, we hypothesized that American Indian and European American women would differ in conceptualizations of domestic violence in terms of definition, historical occurrence, and methods of elimination. In addition, they would differ in terms of their approach to violence from both a cultural prospective and as a reaction to immediate threat (e.g., in terms of the ATVS). American Indian women would demonstrate higher scores on both ATVS subscales, compared to European American women. We also predicted conceptualizations of self with regard to domestic violence would differ (e.g., in terms of the BWS) between the two samples.

Materials and Methods


Twenty American Indian women and twenty European American women from a local community in a medium sized urban area participated in the study. A broad based approach was used for participant recruitment. Information on the study was disseminated to the community (local social service centers and businesses) and potential participants contacted one of the investigators. Previous participants referred additional participants for both groups. The administration at the local Indian Center also publicized the study. Flyers and informational handouts were dispersed around the community to publicize and recruit for the study. This, then, constituted a snowball method of recruitment. Interested individuals could sign up for the study by putting their name on the flyers located in designated places or contacting the researcher directly to set up a time for an interview.

It should be noted that the use of a community sample was crucial for gaining accurate information regarding notions of domestic violence. One issue in gauging accurate prevalence of domestic violence is the use of telephones for gathering information. Reliance on telephones for national family violence surveys may leave out a sizeable number of people who have no telephone access, due to poverty and remote residential locations. It has been estimated that nearly 30% of reservation homes, for example, have no telephone access (U.S. Government Accounting Office 2006). Similarly, poor reading skills may disrupt the accuracy of mail surveys, when trying to reach those without phone services (Chester et al. 1994). Consequently, in order to incorporate as many perspectives as possible for both groups, we relied on a community sample with face to face interviews to provide information concerning definitions and notions of causation regarding domestic violence. We believe this allowed for the most accurate assessment of domestic violence conceptualizations, by allowing those without telephones or with low reading skills to participate.

The tribal affiliations of the American Indian sample were diverse, since over 10 tribes were represented within the sample. There were Ponca of Nebraska (3), Omaha (3), Navajo (1), Rosebud Sioux (1), Northern Cheyenne (1), Santee Dakota (1), Oglala Lakota (2), Winnebago Tribe of Nebraska (1), Chippewa (1), Crow (3), and Ojibwa (1) Sioux (2). For both groups, the mean age was 36 (American Indian M age = 35.60, European American M age = 35.70). Demographic information shows the two groups were representative of the local community, as the 2000 census data for the state lists the female mean age as 36. Information was not collected on income or career oriented questions, because it would have been culturally inappropriate to ask such income questions at one research meeting for the American Indian sample.

Of the American Indian women, 13 were single, 7 were married, and none were divorced, but for the European American sample, 7 were single, 6 were married, and 7 were divorced, X2 (2) = 8.88, p = 0.01. The American Indian women tended to have more children overall (M = 2.40) than the European American women (M = 0.90), F (1, 38) = 10.88, p = 0.002, Mse = 2.07. However, this difference is consistent with national census information (see U. S. Census Bureau 2006).


Participants reviewed and signed a written consent form which outlined the study. After signing the consent form, participants completed a questionnaire concerning domestic violence and were asked semi-structured questions about their perceptions. Both open-ended and rating scale questions were used during the interview. Table 1 contains the items from the interview.
Table 1

Interview items and rating scale items

(Tribal affiliation)

Any children?






1. What is domestic violence?

2. What actions would you consider part of domestic violence?

By the man?

By the woman?

3. What would be the difference between domestic violence and just an argument?

4. What actions would occur so you would call the police?

5. What are the top 5 traits of a victim of abuse?

6. What are the top 5 traits of an abuser?

7. In general, what would you do to end domestic violence?

8. If you or someone you knew were to be involved in domestic violence, what would you do to end domestic violence?

What steps would you take?

9. Do you think children are effected by domestic violence? If so, how?

10. Historically, how to you think your ancestors handled domestic violence?

11. Do you believe the system is fair in how it handles domestic violence?

12. Do you think the system works in how it handles domestic violence?

13. Who do you think starts domestic violence more, men or women, or both?

14. Who should be punished more for domestic violence: Men, women, or both?

15. How much do you believe domestic violence is a problem in your community?

1 2 3 4 5 6 7 8 9

Not at all

Very much so

16. How much do you believe domestic violence is a problem in other communities?

Which one?

1 2 3 4 5 6 7 8 9

Not at all

Very much so

Which one?

1 2 3 4 5 6 7 8 9

Not at all

Very much so

17. What percentage of men experience domestic violence? (for example, 10, 20, 30, 40, 50%)

18. What percentage of women experience domestic violence? (for example, 10, 20, 30, 40, 50%)

19. Have you ever experienced domestic violence?

If yes, what did you do about it, if anything?

If no, what do you think keeps you from experiencing it?

20. What do you think causes domestic violence?

In addition, the BWS (Schwartz and Mattley 1993) and the ATVS: Revised (Funk et al. 1999) were completed to determine if attitudinal differences existed concerning battering and violence. The BWS is composed of 22 items that reflect notions of emotionality, security, competence, agentic responding, and forcefulness. The items were rated on a scale with 0 (strongly agree) to 4 (strongly disagree) as endpoints. Total scores could range from 0 to 88.

The ATVS was originally composed of 22 items, and all these items were included for completion. Items were rated on a 5 point rating scale (1 = strongly disagree to 5 = strongly agree). The Culture of Violence subscale is composed of six items, and the Reactive Violence subscale is composed of seven items. Only findings for the subscales were analyzed.


All research interviews were conducted in a medium sized urban community. In order to interview Indian women in a culturally appropriate, non-threatening environment, Indian women were given the option to hold their interview at the local Indian Center in a private setting or to choose a convenient location (i.e., place of employment, the university, residence, etc.). Interviews with American Indian women were conducted by an American Indian interviewer (one of the investigators). Some respondents moved between the urban community and a reservation or had frequent contact with reservation life. The European American women could choose the location that they felt was most convenient and private as well (i.e., place of employment, the university, residence, etc.).

At the time of data collection, an investigator read the consent form to participants, and participants read and signed the consent form. The consent form explained the nature of the study and what information would be requested, including the fact that withdrawal from the study at any time was possible. After completion of the interview and attitude questionnaires, participants were debriefed and told of hypotheses. Time was taken to ensure that all participants’ questions were answered. In addition, a form with the name and phone numbers of local shelters and counseling centers was provided to ensure that participants gained information concerning local resources on abuse issues. Contact information for the interviewers was also supplied.

For the open-ended responses, two raters, who were blind to the hypotheses and ethnic condition, sorted the answers into common categories. This process enabled assessment of inter-rater reliability. The inter-rater reliability was 97%. Any differences that occurred in the categorizations were resolved by one of the investigators. A series of X2 analyses (Pearson Chi-Square statistics) and analyses of variance were used.


Figure 1 indicates American Indian women tended to believe physical acts of violence defined domestic violence, and European American women tended to define emotional and verbal acts as domestic violence, X2 (1) = 12.38, p < 0.001.
Fig. 1

Frequency of responses for “What is domestic violence?” by ethnic group. AI, American Indian; EA, European American; Phys Acts, physical acts of violence; Ver/Emo/Phys, emotional, verbal, and physical acts of violence. X2 (2) = 12.38, p = 0.001

As shown in Fig. 2, in terms of the actions of domestic violence, a significant difference was found, such that American Indian women felt that domestic violence contained an argument and European American women did not, X2 (1) = 11.65, p = 0.001.
Fig. 2

Frequency of responses to “What actions would you consider part of domestic violence?” by ethnic group. AI, American Indian; EA, European American; No, no mention of physical violence; Yes, mention of physical violence. X2 (1) = 11.65, p = 0.001

Table 2 displays the responses concerning what men and women do as part of domestic violence. We analyzed the responses for men’s and women’s behaviors with separate X2 tests. American Indian women viewed a man’s physical actions as constituting domestic violence, but European American women included verbal and emotional abuse by a man as domestic violence, X2 (2) = 15.10, p = 0.001. In contrast, American Indian women tended to believe domestic violence was present when women showed physical violence, while European American women believed it was present when women showed physical or verbal abuse, X2 (5) = 13.79, p = 0.02. It should be noted that American Indian women also mentioned that women would defend themselves, put up with abuse, and live in fear as part of the actions occurring with domestic violence. Several American Indian women mentioned physical retaliation as an appropriate response to abuse. European American women mentioned none of these.
Table 2

Frequency of responses for perceived female and male actions considered part of domestic violence by ethnic group


Ethnic group

American Indian

European American

Actor sex





Physical acts















Control life










Puts up with










No response





Ver/Emo/Phys: verbal, emotional and physical abuse; Ver/Phys: verbal and physical abuse; Defend: defensive behaviors; n = 20 for both ethnic groups

While both groups believed abusers were controlling (American Indian n = 7, European American n = 13) and short tempered (American Indian n = 4, European American n = 5), there were differences in perceived traits. American Indian women mentioned that abusers were manipulative (n = 5) and had a history of abuse, experienced victimization themselves, and were either drunk or on drugs (n = 6). European American women did not mention victimization or family history.

While both groups of women believed abuse victims were isolated and alone (American Indian n = 7, European American n = 12) and possessed low self esteem (American Indian n = 8, European American n = 10), American Indian women mentioned familial history (e.g., past history of abuse, father was abusive, n = 6), but European American women did not. American Indian women mentioned that abuse victims experienced shame and were ashamed (n = 5), but European American women did not. European American women focused on the abuse victim’s use of excuses for the abuser (n = 9), apologizing for the abuser (n = 3), and pretending (n = 4).

As indicated in Fig. 3, American Indian women tended to believe the only difference between an argument and domestic violence was the occurrence of physical violence, while European American women believed the difference occurs with emotional, verbal or physical abuse, X2 (2) = 23.75, p = 0.001.
Fig. 3

Frequency of responses to “What would be the difference between domestic violence and just an argument?” by ethnic group. AI, American Indian; EA, European American; Phys, physical acts of violence; Ver/Phys, verbal and physical acts of violence; same, the two are the same. X2 = 23.75 (2), p = 0.001

In terms of causation, European American women tended to say that domestic violence was caused by personal, internal dysfunctions of the abuser, such as anger control issues (Fig. 4), whereas American Indian women viewed the cause of domestic violence to emanate from society and social problems, such as poverty, unemployment, and lack of mobility due to isolation, X2 (3) = 21.03, p < 0.001. Thus, as predicted European American women focused on internal attributions for domestic violence, while American Indian women focused on external causes.
Fig. 4

Frequency of responses to “What causes domestic violence?” by ethnic group. AI, American Indian; EA, European American; Soc and Soc Prob, society and social problems; Upbringing, actors’ upbringing, Per Dys, personal dysfunctions, X2 (3) = 21.03, p = 0.001

In response to how domestic violence was handled historically, European American women believed their ancestors hid domestic violence and never talked about it, whereas American Indian women stated domestic violence did not occur in their community historically, X2 (3) = 25.05, p < 0.001, as shown in Fig. 5.
Fig. 5

Frequency of responses to “Historically, how do you think your ancestors handled domestic violence?” by ethnic group. AI, American Indian; EA, European American; Hid it, they hid the problem; With Vio, handled it with violence; Didn’t Hap, domestic violence didn’t happen. X2 (3) = 25.05, p = 0.001

When asked if the legal system worked when dealing with domestic violence, European American women were more likely to state that it worked some of the time (Fig. 6), and American Indian women were likely to state that it did not work, X2 (3) = 9.03, p = 0.03. Although differences in perceptions of the fairness of the legal system were not significant, American Indian women tended to believe the system was not fair (n = 14), while European American women tended to believed the system was fair sometimes (n = 9), p = 0.15.
Fig. 6

Frequency of responses to “Do you think the legal system works in how it handles domestic violence?” by ethnic group. AI, American Indian; EA, European American; No, System does not work; Yes, System does work; Sometimes, System works sometimes. X2 (2) = 9.03, p = 0.03

As shown in Fig. 7, with regard to what would have to occur in domestic violence in order for them to call the police, almost 3/4 of the American Indian women stated that physical violence or extreme physical violence, requiring medication attention, would need to occur before they would call the police. However, over half of the European American women stated they would call the police after threats were made, before the situation escalated, X2 (3) = 11.20, p = 0.01.
Fig. 7

Frequency of responses to “What actions would occur so you would call the police?” by ethnic group. AI, American Indian; EA, European American; Phys, physical violence; Extreme Injury, extreme physical injury. X2 (3) = 11.20, p = 0.01

All the European American women (n = 20) stated both men and women should be punished for domestic violence, while American Indian women’s responses were more varied. While most believed that both should be punished (n = 12), some believed men should be punished more (n = 6) and a few believed women should be punished more (n = 2), X2 (2) = 10.00, p = 0.01 as shown in Fig. 8.
Fig. 8

Frequency of responses to “Who should be punished more for domestic violence: men, women, or both?” by ethnic group. AI, American Indian; EA, European American. X2 (2) = 10.00, p = 0.01

The women were asked to rate how much of a problem domestic violence is in their communities on a 9 point rating scale (1 = not at all and 9 = very much so). American Indian women indicated that domestic violence was a larger problem (M = 8.00, SD = 1.30), compared to European American women (M = 6.15, SD = 1.73), F (1, 38) = 14.69, p < 0.001, Mse = 2.33.

When asked what percentage of women experience domestic violence, American Indian women stated that nearly 69% (M = 68.55) experienced domestic violence, while European American women believed 47% experienced such abuse (M = 47.20), F (1, 38) = 8.73, p = 0.005, Mse = 522.11. However, the women did not differ, statistically, on an estimated percentage of men they believed experienced abuse in their communities (American Indian women M = 35.20%, European American women M = 26.30%), F (1, 38) = 1.57, p = 0.22. Moreover, they did not differ, statistically, in their own experience with domestic violence, although American Indian women reported more personal experience (n = 18), compared to European American women (n = 14), p = 0.11.

As predicted, American Indian women had a higher rating on the total ATVS (M = 33.80, SD = 9.42), compared to European American women (M = 28.30, SD = 8.10), but the difference was marginally significant, F (1, 38) = 3.92, p = 0.06, Mse = 77.20. The Chronbach’s alpha reliability coefficient was 0.79. The Chronbach’s alpha reliability was consistent across cultures (American Indian = 0.77; European American = 0.80), as well.

The two subscales were highly correlated (r2 = 0.71, p = 0.001); consequently, a multivariate analysis of variance test for the subscales by group was conducted and was marginally significant, F (2, 37) = 2.90, p = 0.07. The Culture of Violence subscale showed a marginally significant difference, F (1, 38) = 3.62, p = 0.065, Mse = 13.36, with American Indian women scoring higher on the scale (M = 10.55, SD = 3.91), compared to European American women (M = 8.35, SD = 3.38), as predicted.

The Reactive Violence subscale indicated a significant difference between the two groups, F (1, 38) = 5.83, p = 0.02, Mse = 36.30. American Indian women had more general acceptance of reactive violence (M = 17.95, SD = 7.19) than did European American women (M = 13.35, SD = 5.58). To evaluate differences between the two groups of women on the seven individual items from the scale, independent-samples t tests were conducted on the Reactive Violence subscale items. Results of the significant differences can be found in Table 3. Items such as “if a person hits you, you should hit them back” and “it’s okay to beat up a person for badmouthing me or my family” produced significant differences between the groups. European American women were less likely to endorse the two statements, compared to American Indian women. American Indian women were less likely to endorse the statement, “if someone tries to start a fight with you, you should walk away”, compared to European American women.
Table 3

Mean differences between American Indian and European American women on culture of violence and reactive violence subscales by ethnic group

Ethnic Group

Reactive violence items




\( \eta ^{2}_{p} \)

If a person hits you, you should hit them back





It’s okay to beat up a person for badmouthing me or my family





aIf someone tries to start a fight with you, you should walk away





AI, American Indian; EA, European American

aItem was reverse scored

**p < 0.01

On the BWS, the Chronbach’s alpha reliability coefficient = 0.82. American Indian women had a lower mean score (M = 42.00, SD = 9.80), compared to the European American women (M = 50.75, SD = 12.03), F (1, 38) = 6.36, p = 0.02, Mse = 120.36, which demonstrates a general self conceptualization that does not fit a battered woman who is lacking agency and forcefulness.

It could be that differences in perceptions of what constitutes domestic violence were not due to ethnic group, but were due to differences in attitudes on the BWS and ATVS. We conducted a simultaneous entry multiple regression with “What is domestic violence?” as the dependent variable and with group, BWS total score, Culture of Violence and Reactive Violence scores as the independent variables. The overall analysis of variance was significant, F (4, 35) = 4.68, p = 0.01. Of the independent variables, only ethnic group significantly influenced the notion of domestic violence (β = −0.47, SE = 0.15, p = 0.01), while Culture of Violence (β = 0.19, SE = 0.03, p = 0.37), Reactive Violence (β = −0.25, SE = 0.02, p = 0.22), and BWS (β = 0.12, SE = 0.01, p = 0.45) were not significant predictors in the model.

In addition to the codified, open-ended responses and the attitudinal measures, American Indian women offered comments concerning domestic violence that were not part of the actual interview process. These responses indicated the occurrence of extreme violence in their lives and the responses to violence they would take. For example, several American Indian women mentioned murder and death as outcomes of domestic violence, and some mentioned women they knew who were in jail for defending themselves against domestic abuse. They reported the use of retaliation as a possible response to domestic abuse. European American women never mentioned such information, even with additional inquiries about such events.

Both groups believed that men were more likely to start domestic violence (American Indian women n = 14, European American women n = 14, p = 0.58) and both groups reported similar methods to end domestic abuse. ‘Get out of the situation’ was the most frequent response (American Indian women n = 10, European American women n = 14, p = 0.56). However, American Indian women were more likely to also refer to spiritual healing and traditional education (n = 7).

The two groups of women did not differ on the perceived effects of domestic violence on children. Both groups mentioned that trauma was the result of exposure to such violence. They both mentioned that domestic violence had intergenerational effects on children’s behavior.


As hypothesized, American Indian women’s perceptions of domestic violence differed from that of European American women, in terms of definition, historical occurrence, causation, current rates, and the effectiveness of the legal system. No differences emerged in terms of the rate of men who are involved, the likelihood of men starting abuse, methods to end domestic violence, or the ramifications for children.

American Indian women tended to emphasize physical abuse as domestic violence, whereas European American women tended to define verbal and emotional acts along with physical abuse as domestic violence. This confirmed our notion that American Indian women would focus on actions, while European American women would focus on internally driven motivations (emotional and verbal expressions) and actions to conceptualize abuse. Thus, American Indian women’s definition infers a higher threshold of violence needs to occur to be acknowledged as domestic abuse, and this was substantiated by the reported actions needed to contact police, and differences in ATVS scores as well.

The ATVS showed a marginally more pro-violence attitude on the part of American Indian women. As the extremity of domestic violence seems to be greater for American Indians, higher scores for American Indian women on overall attitudes towards violence may indicate one ramification of belonging to a group where chances of violence are high, since none of our American Indian women sample reported that they were currently in abusive situations. In addition, American Indian women reinforced oral and written accounts of the lack of domestic violence in historical and traditional Indian family life. While American Indian women believed domestic violence was rare, European American women believed it was hidden and not discussed. Thus, while historical incidents were lacking, American Indian women have psychologically adapted to a violent world with more lenient reactive violence attitudes, and an acknowledgement of the intensity of anticipated domestic abuse for themselves and others.

In women’s conceptualizations of self, “race is an important factor which acts as a cultural variable” (Schwartz and Mattley 1993, p. 285), and this may have implications for the BWS. Portman (2001) found American Indian women tended to score higher on masculine subscales than European American women, when examining sex role attributions, but found no difference between the two groups on the feminine subscale. “It might be ascertained that American Indian women do not fit the dominant US society norms for traditional sex roles” (p. 80) in that they “embrace” femininity, but also embrace masculine traits. Our results seem to compliment the Portman findings. The American Indian women sample did score higher on notions of agency and forcefulness captured in the BWS, compared to the European American sample. Future research should investigate whether current conceptualizations of self and sex roles differ between American Indian women and other ethnic groups.

Several factors negate most dominant culture methods to reduce and/or end domestic violence in Indian Country (i.e., legal jurisdiction, access to health education programs, culturally sensitive counseling, etc.). In a recent survey of Indian Health Services (IHS), only 62% of facilities reported screening for domestic violence, with higher screening rates reported by hospitals and facilities with domestic violence policies and procedures, and lower rates in clinics that were tribally administered, rather than IHS administered (Clark 2001). Bohn (1998) proposed that response programs incorporate women’s spirituality and culture and allow women to reclaim their self and strength. Elders, healers, teachers, and counselors should be involved in the healing process. Values, like dignity and respect should be reinforced. Taken together, the findings here provide insight into program development and implementation when working with American Indian women.

The development and implementation of tribally controlled and culturally appropriate domestic violence shelters and counseling seems imperative. Programs such as Cangleska, White Buffalo Calf Women’s Society, and Ama Doo Alchini Bighan, are examples of tribally developed responses to domestic violence. The United States Department of Justice has made funding for tribally developed and controlled shelters a priority, in hopes of increasing responsiveness to American Indian women and reducing the prevalence of violence for American Indian communities. Resources and strengths of communities should be mounted to respond to abuse, and the history of oppression and domination within communities must be considered (Hamby 2000; Willis Esqueda and Tehee 2005). It is essential that responses be generated within communities, rather than imposed by outsiders, particularly given the unique differences in preference for legal system intervention and notions of effectiveness found in our samples.

The findings in this research provide information that can assist tribally controlled, federally controlled, and urban area health centers to better understand and educate against domestic violence within Indian country. Although tribally developed and controlled shelters are imperative, the majority of Indian women live in urban areas where they can become invisible in the service mix offered in the local urban community. Our findings indicate educational programs should include a component addressing American Indian women’s conceptions and needs with regard to domestic violence. Emergency rooms, treatment programs, and shelters should be sensitized to American Indian women’s unique cultural and psychological approaches to violence, as evident from the findings reported here.

There is a dearth of research examining violence as it impacts American Indian women (Lujan 1992). Future research might address the origins of difference between American Indian and European American women in their domestic violence assessments. We should determine if differences lie in adoption of biased stereotypes by both groups about their in-group and violence. For example, it may be that American Indian women believe American Indian women, as a group, are stronger and can withstand more than the typical woman, while European American women may believe their in-group is less strong and in need of rescue from violence. It may be the lack of intervention by the community and the legal system has rendered American Indian women leery of any outside assistance, while European American women expect such intervention, because domestic abuse has been an historical social problem in their community.

In addition, differences in American Indian and European American women’s conceptualizations of violence may be created by the prevalence rates of violence in their daily lives. Although actual rates of violence do differ between American Indian and European American communities, perceptions of domestic abuse prevalence in one’s community may undermine psychological health. Thus, future research should address whether perceived community prevalence rates influence definitions and attitudes towards self and violence regardless of ethnic group. “We have to stop this current before it floods our very existence,” (Hedge Coke 1998, p.326).


The use of labels, such as American Indian, Native American, Indian, and indigenous people, are used interchangeably throughout the manuscript. Although it is recognized that there are conceptual and political issues tied to these labels, this usage is not meant to offend any individual or group. These terms reflect the numerous labels that are currently used in and among American Indian communities, and within the social sciences. Further, it is recognized that such terms can apply to all indigenous people within the Americas; however, the focus here is on North American Indians.


Copyright information

© Springer Science+Business Media, LLC 2007