Gender Role Attitudes and Attribution of Blame for Spousal Violence in Married Men and Women

Sobia Masood and Anila Kamal, Ph.D

National Institute of psychology,

Quaid-i-Azam University, Islamabad.

The present research aimed at investigating the effects of gender role attitudes on attribution of blame for spousal violence in married men and women. Spousal Violence Blame Questionnaire (SVBQ) was developed which was based on 12 visual scenarios of spousal violence incidents; scenario opinion questionnaire; and background information/story cards. A sample of 120 married participants (60 men, 60 women) with age ranging from 20 to 62 years (M= 35.63) was administered by Gender Role Attitude Scale (Kamal & Saqib, 2004), Dyadic Adjustment Scale (Naseer, 2000) and SVBQ. Experimental group was given the background story before presenting video clips, whereas the control group was shown the clips without background story. Experimental group attributed significantly more blame towards the victim in a spousal violence incident as compared to the control group. Participants considered an incident as spousal violence in case of physical violence scenarios but they did not take it as violence in case of non physical violence.

Keywords: Attitudes, Attribution, Blame, Spousal Violence

A sample of 120 married people (60 men, 60 women) with age ranging from 20 to 62 years (M= 35.63) was administered by Gender Role Attitude Scale (Kamal & Saqib, 2004), Dyadic Adjustment Scale (Naseer, 2000) and SVBQ. Experimental group was given the background story before .presenting video clips, whereas the control group was shown the clips without background story. Experimental group attributed significantly more blame towards the victim in a spousal violence incident as compared to the control group. Participants considered an incident as spousal violence in case of physical violence scenarios but they did not take it as violence in case of non physical violence.

The term spousal violence includes physical, sexual, psychological abuse and stalking committed by one spouse against the other in a marital relationship. Spousal abuse is physically or sexually assaulting or psychologically abusive behavior by a 'partner against the. other in an intimate, sexual, peer and usually cohabitating relationship (Ganley, 1982).

As proposed by Kelley (1972), "Attribution processes are to be understood, not only as means of providing the individual with a veridical view of his world, but as means of encouraging and maintaining his effective exercise of control in that world". According to him, people's reactions to victims are affected by their desire to avoid blame for their own future accidents. According to Chaikin & Darley, people make defensive attributions and assign causality in order to maintain or enhance their self esteem (as cited in Kelley, 1972).

Almost all research concerning domestic violence has found an effect of gender. In comparison to men, women are less likely to blame the female victim and more likely to attribute responsibility for the incident to the male abuser (Harris & Cook; Pierce & Harris as cited in Locke & Richman, 1999). Harris and Cook (as cited in Locke & Richman, 1999) found that women sympathized more with the victim, regardless of similarity between the victim and participant. Kristiansen and Giulietti were of the view that whether male or female, as attitudes toward women become more positive, people tend to blame the abuser more than the victim (as cited in Locke & Richman, 1999).

Tang and Tam (2001) has cited the work of many authors according to which people who endorse traditional attitudes toward women and their appropriate gender roles in the society also tend to endorse violence against women-related myths and to blame the victims for their victimization rather than to hold the perpetrators responsible for their violence. In addition, perceived prevalence of violence against women and its negative effects on the victims and community have also been found to influence people's judgments of responsibility, which in turn affect referral, treatment, and rehabilitation decisions for the victims and perpetrators.

Demographic characteristics of age, gender, and educational attainment are related to the assignment of responsibility to domestic violence situations. Despite some mixed results, most studies cited by Tang and Tam (2001) have shown that women tend to assign less responsibility to rape victims than men. Schneider asserted that older people, as compared to their younger counterparts, tend to hold conservative views about women's roles in the society and assign more responsiblility to female victims of domestic violence (as cited in Tang & Tam, 2001). According to Allison & Wrightsman, as education can provide people with broader perspectives on various human behaviors, those with a higher educational attainment are found to have more liberal views about women and to hold domestic violence victims less responsibility for their misfortune (as cited in Tang & Tam, 2001).

The above factors are evident in public service professionals' assignment of responsibility to domestic violence victims and perpetrators. Research has shown that victim-blaming attitudes and behaviors among police officers and medical personnel are associated with traditional views of women, endorsement of domestic violence - related myths, and poor knowledge about the effects of domestic violence on the victims (as cited in Tang & Tam, 2001).

According to Acock and Ireland, people with more traditional gender role attitudes have been found to attribute more responsibility to the victim of rape scenario than those who hold less stereotypical and more progressive beliefs about gender. Also traditional sex role attitudes have been found to be related to victim blame for attitude condoning marital rape (as cited in Pollard, 1992).

According to Stormo, Lang, and Stritzke (1997), abuser men assigned more blame to their victims than to themselves, whereas non- abuser men assigned more blame to the personality of the abuser than to other factors. In a case where an equal level of alcohol was consumed by both victim and abuser, the victim will be attributed more blame than the sober victim. This study found that overall, males were assigned more responsibility and blame for the domestic violence than females.

Moore (1998) cited the work of many researchers, according to these studies, sexist attitudes are based on stereotypical views of gender appropriate behavior. Tolerance of spousal violence has been correlated with sexist ideology, hostility towards women, acceptance of interpersonal violence, adversarial sexual beliefs, and rape myth acceptance These attitudes are based on stereotypical beliefs that the male has the prerogative to initiate aggression and to use some pressure to control others. Sexist attitudes were expected to add to the prediction of blame of the target of domestic violence in this study.

Ward (1995) reported the work of Krulewitz & Payne that some studies have shown that women attributed more responsibility to victims, at least under certain circumstances. However, men appear to attribute more blame and responsibility to victims on the basis of their characters than do women (Ward, 1995). More specifically, men are more likely to perceive a women victim as contributing to the spousal violence incident

Liberal gender attitudes are the most salient predictors of a low level of responsibility attribution, to domestic violence victims, wher,eas perceived serious negative effects of domestic violence on the victims are the most salient predictor of a high level of responsibility attribution to domestic violence perpetrators (Tang &Yan, 2003).

There are several reasons for conducting this research. In Pakistan not much research has been conducted on attribution of blame for spousal violence. In review of the existing research on the issues of spousal abuse and family violence, Huisman stated that it is apparent that the focus of the research on violence against women has been on the western culture, specifically, white, middle-class women (as cited in Moore, 1998). As many researchers have pointed out, the nature of the violence and attribution of the causes, blame for spousal violence and abuse in Asian families require special attention (Ho, 1990). The issue of family violence in Pakistan is highlighted these days by different NOOs (e.g., Rozen, Sach, Shirkat Gah), but there is very little proper research work available on this issue.

Due to the continuing problem of spousal violence in Pakistani society, it is important to understand factors that may influence attributions concerning domestic assaults. The present research seeks to do this y examining gender-role attitudes on domestic violence attributions. Specifically this research investigates whether victim blame and derogation increase when people having traditional gender role attitudes attribute blame for spousal violence incidence

In two experimental conditions, attribution of blame for spousal violence would be measured using video clippings bas d on scenarios of spousal violence. Such scenarios have been used in various experimental studies (Moore, 1998; Valentine-French & Radtke, 1993; Walster, 1966).The results of the present study are expected to help learn the mediating role of gender role attitudes in attribution of blame for spousal violence. The use of experimental method for this study is significant as it involves direct intervention on the part of the researcher. As social behavior is complex and more than one variable cause certain effect on the given behavior. Therefore, to see the effects on attribution of blame for domestic violence, more than one variable would be taken at a time for the current study.

Objectives

The goal of the present research is to investigate the relationship between gender role attitudes and attribution of blame for spousal violence. Moreover, the research also explores the compatibility between perceived gender role attitudes of respondent and his/her attribution of blame towards spousal violence. Furthermore research also plans to meet the following objectives:

To develop a scale for the measurement of attribution of blame for spousal violence based on video based scenarios.
To find out the differences between men and women on their gender role attitudes, and attribution of blame for spousal violence.
To see the attribution of blame patterns under different scenarios i.e., attribution of blame in physical violence incidents, nonphysical violence incidents, man abuser scenarios, women abuser scenarios etc.

Hypotheses

To meet the objectives of the present research, following hypotheses were formulated:

Women will have more modem gender role attitudes than men. .
Participants will be more likely to label the scenarios based on physical violence as 'spousal violence' than the scenarios based on nonphysical violence.
Participants from experimental group will assign more blame towards the victim of spousal violence as compared to control group.

Method

Research Design

The present research was completed in 3 parts. Part I of the research deals with (i) the development of questionnaire of attribution of blame for spousal violence based on video clippings, and (ii) development of statements (brief background information about the scenarios). Part IT is the experiment-I (pilot study) and is concerned with the determination of reliability of the dependant measure and to see the effects of the brief background information statements on the attribution of blame. Part III is experiment-II (main study) intended to explore the relationship and effect of attribution of blame for spousal violence and gender role attitudes of married men and women. Gender differences in Attribution were also explored by categorizing participants according to traditional or modem gender role attitudes, using Gender Role Attitude Scale (Kamal & Saqib, 2004), and measuring their attribution of blame in two different experimental conditions. Dyadic Adjustment scale (Naseer, 2000) was also used to screen the participants for final analysis and only' those participants were included in the final analysis that had high level of marital adjustment.

Experimental Design

This study employed 2x2x2x2 factorial design to compose 12 spousal violence video clips. The four factors examined in the study are the (1) abuser's gender (man vs woman), (2) type of the violence (physical vs non physical), (3) experimental conditions (giving background story based on gender role attitudes vs not giving any background story regarding approaching scenario), and (4) gender role attitudes (traditional vs modem). To facilitate the process of attributing blame for the study participants, the present study composed 12 "ambiguous" video clips of spousal violence incidents by including two main types of spousal violence i.e., physical violence and non physical violence (includes emotional/psychological, social and economic violence). Only sexual violence is not included in selected scenes as the judges in the phase III of Part I of the study highlighted the point of cultural limitations. They suggested not adding scenes showing sexual violence because in Pakistani society, people might have hesitation while watching those scenarios. All of the 12 clips included following four elements (1) The perpetrator and the victim are spouses (2) the perpetrator can be anyone, the husband or the wife (3) no weapon was involved, (4), clip was showing gender role attitudes.

Part I of the Research

Attribution of the blame for spousal violence was measured by using a scale based video clippings of scenarios on spousal violence. This part was completed in four phases. Phase I: Identification of different spousal violence types Phase II: Gathering movie clipping based on spousal violence incidents. Phase III: Final selection of movie clippings according to their appropriateness, and development of scenario opinion questionnaire. Phase IV: development of statements (brief background information about the scenarios so that experimental conditions would be manipulated).

Part II of the Research

Part II of the research was the experiment-I purported to see the difference between experimental and control group. A sample of 40 (20 men, 20 women) married individuals were administered the GRAS (Kamal & Saqib, 2004), and Dyadic Adjustment Scale (Naseer, 2000) along with the SVBQ. Finding revealed a significant difference between the experimental and control group confirming the effect of the background information/story of the scenarios on attribution of the blame for spousal violence.

p align="center"> Part III of the Research

Participants

A sample of 120 (60 men, 60 women) married participants from different areas of Rawalpindi/Islamabad, Gujar Khan, Kohat and Peshawar was selected on volunteer basis. Participants were selected by using purposive sampling technique.

Inclusion criteria

Only those participants were included who had been married for at least 3 years, had high marital adjustment, had at least one child, not used to regularly watch movies (view not more than 15 movies per year) and those who were at least matriculate. This was done to control extraneous and confounding variables like effect of previous exposure of scenarios on the participant's attribution, effect of their own marital maladjustment on their attribution of blame in spousal relationships etc. Participants had age ranging from 20 to 62 years (M = 35.63; SD = 7.44). About 35 % participants had an education level up to F.A and remaining 65% were B.A and above. Majority of the participants did not watch movies at all (65.8%), whereas 6.7% participants watched less than 5 movies per year and only 27.5% participants watched 5 to 15 movies per year.

Measures

Gender Role Attitude Scale (GRAS).

Sex Role Attitude Scale was originally developed by Kamal and Ansari in 1992 and was revised by Kamal and Saqib (2004) and the word 'sex' was also replaced with 'gender' in the title of the scale along with the other revisions of the scale. GRAS is a 30 item scale which assesses the- attitudes regarding roles of men and women inside and outside their homes; parental responsibilities of men and women; occupational abilities of men and women; vital life decisions; personal relationship between. men and women; and level and type of academic achievement for men and women. It's a 5-point rating scale on which respondents have to express his/her degree of agreement or disagreement. Half the items are phrased in traditional manner and half in modem. Scoring for traditional items is reversed so that the high total score indicates modem views. The modem (positively phrased) item numbers are 2, 4, 6, 7, 10, 11, 12, 13, 18, 19, 20,22,23,26,27, and the traditional (negatively phrased) items numbers are: 1, 3, 5, 8, 9, 14, 15, 16, l7, 21, 24, 25, 28, 29, 30 . The reliability of the scale is reported to be .81 by the authors.

Dyadic Adjustment Scale (DAS).

Dyadic Adjustment Scale was originally developed by Spanier (1976) to measure marital adjustment. It was adapted and translated into Urdu by Naseer (2000). In the present study, the adapted and translated version was used. It consists of 26 items. Item no. 1-20, 22, 23, 24, 25 were responded on a 6-point rating scale. Item no. 21 was rated on a 5 point rating scale with 0 standing for 'not at all', and 4 standing for 'in almost all'. In item no. 1-13 and 16-17, the response category 'always agree' was assigned a score of 5 and 'always disagree' was assigned a score of O. Item no. 14, 15, 18, 19, and 20 were negative items and were reverse scored. Item no. 22, 23, 24, and 25 were scored with 5 standing for 'frequently' and 0 standing for 'never'. Item no. 26 was dichotomous with 1 standing for 'Yes' and 2 standing for 'No'. Score ranged from 1-124. High score indicate high marital adjustment. The instrument has reliability coefficients of .80 and validity is also satisfactory.

Spousal Violence Blame Questionnaire (SVBQ).

Spousal Violence Blame Questionnaire which measures attribution of blame for spousal violence was developed by the researcher in part-I of this study). This scale consists of (i) 12 video based scenarios; (ii) a scenario opinion questionnaire on which questions regarding each scenario are written; and (iii) Scenario Instruction/Brief Background Information Cards' (i) Scenarios. There are 12 video based scenarios depicting spousal violence incidents (6 scenarios on physical violence, and 6 scenarios on non physical violence). The genders of the aggressor/abuser and target/victim were also varied across clippings. Husband is shown as an abuser in 6 scenarios (3 physical violence, 3 non physical violence) while wife is shown as an abuser in remaining 6 scenarios (3 physical violence, 3 non physical violence). Duration of each movie clipping was approximately of 1 minute. Each movie clip detailed a scenario in which two individuals became involved in a domestic violence situation based on the two main types of domestic/spousal violence (i.e., physical violence, and non physical violence). This interaction culminated in one of the pair acting in an aggressive manner toward the other, either shouting at, or shouting at and then purposely bumping the other person or doing any other abusive act. Before presenting the scenario, participants of experimental group are given some written background information/story of the approaching scenario in a line or two, which was prepared in the part-I of the present study by the researcher. Participants from control group are not given any background information/story regarding the approaching scenario.

(ii) Scenario opinion questionnaire (SOQ).

The present study was designed to measure the attribution of blame toward spousal violence abuser and victim. Scenario opinion questionnaire was based on 5 questions. Question no. 1 and 2 were asked to investigate the judgment of the blame towards abuser and victim and the reasons of attributing the blame. The study asked the participants who is more blameworthy on a dichotomous question with" I" standing for "Yes" and "2" standing for "No" ("In your opinion, who is to be blamed more for this incident?"), and the reasons for her/him being blamed more were asked on an open- ended question ("In your opinion, for what reasons she/she is more blameworthy"?).

Questions no. 3 and 4, a Likert type scale, measuring attribution of blame for each scenario depicting spousal violence incident include: ''To what extent do you think husband should be blamed for the incident"? and "To what extent do you think wife should be blamed for the incident"? In scene No. I, 3, 5, 7, 9, and II, the husband was abuser and in scene No.2, 4, 6, 8, 10, and 12, the wife was shown as an abuser. Participants were asked to indicate their opinion on a 4 point Likert scale, with response categories of 'not at all', 'to some extent', 'to great extent' and 'full blameworthy'. Scoring was done as '1' standing for 'not at all', '2' for 'to some extent', '3' for 'to great extent' and '4' for 'full blameworthy' for abuser blame and reverse scoring was done for-victim blame i.e., 4 standing for 'not at all', '3' standing for 'to some extent' '3' 'for 'to great extent' and '1' for 'full blameworthy'. There were total 24 questions in 12 scenarios and score ranged from 24 to 96. High score indicate high abuser blame and low score indicate high victim blame.

Procedure

2-4 individuals in one group were administered the scales. Between participant design was used in the present study and the participants were randomly divided into two groups to assign two different experimental conditions to them. Written instructions were handed over to each group of potential participants. Half of the participants filled the Gender Role Attitude Scale (GRAS) and Dyadic Adjustment Scale (DAS) before filling the Spousal Violence Blame Questionnaire (SVBQ). The procedure was reversed for the other half people. Experimental group was given a brief background information/story card of the approaching scenario while control group was not given any information regarding the approaching scenario. The administration of all three scales took approximately 90-100 minutes to complete for most of the individuals. Once completed, packets were collected from.

Results

Table 1

Table 1 shows a non- significant difference between men and women on gender role attitudes. The results reject the hypothesis that women will have more modem gander role attitudes than men.

Table 2

Table 2 reveals that there is non-significant difference between men and women on attribution of blame for spousal violence scenarios expect the scenarios of nonphysical violence with women abuser, the difference between men and women is significant.

Table 3

Table 3 shows that people who were given the back ground information regarding approaching scenarios (experimental group) attributed more blame to the victim of spousal violence as compared to those who were not given any background information (control group). Finding reveals a significant difference (p<.Ol) on almost all scenarios of spousal violence except non-physical violence by women abuser. These findings confirm the hypothesis that "participants from experimental group will assign more blame to the victim as compared to control group".

Table 4

Table 4 shows highly significant difference .on'the labeling of scenarios as spousal violence on the sample. In physical violence scenarios, participants labeled the incident as violent but in nonphysical violence scenarios, they did not label it as violent. The findings revealed a great deal of disagreement and confusion about how to label the spousal violence incident. When the participants were shown scenarios based on physical violence, more participants labeled it as 'spousal violence', but when they were shown the scenarios based on nonphysical violence, less participant labeled it as spousal violence.

Discussion

In the study of gender differences on attribution of blame for spousal violence scenarios and gender role attitudes of the participants, it was assumed that women will have more medium gender role attitudes than men and will attribute significantly more blame to the abuser in spousal violence scenario as compared to men. But surprisingly, the findings rejected both of these hypotheses and revealed that there is no significant difference between men and women on GRAS and SVBQ. Even, the mean scores of GRAS indicate that neither men nor women have more egalitarian gender role attitudes, and women have slightly more traditional gender roles than men which is a very interesting finding of the present research. In Pakistani society gender role ideology that a man should be a breadwinner and a woman should be a housekeeper exists quite obviously. In the daily life, we see that men of our society have more traditional views (typical patriarchal society) than women. It also has been supported by a number of researches (see, for ex.ample, Abouchedid, Marzillier, & Hall as cited in Salik, 2003). On the other hand, findings of the present research suggest that there is no difference between men and women on gender role attitudes which are in accord with Barbara (2003) who also found that both men and women had traditional gender role attitudes. Brien and Fassinger (as cited in Luo, 1999) pointed that women's traditional gender role attitudes may be a result of lower socio economic circumstances.

As the attribution of blame is concerned, there is no significant difference between men and women on attribution of blame for spousal violence scenarios, except the scenarios of nonphysical violence with women abuser, the difference between men and women is significant. Many studies have shown that men are more likely than women to attribute blame to both men and women victims (Kenig & Ryan, 1986; Rubin & Borgers, 1990). The findings of the present research are not consistent with previous researches. However, the slight difference between the means scores of men and women suggest that women put more blame towards abuser, and men attributed more blame to the victim of spousal violence. Burt (1980) assumed that patriarchy shapes attitudes and beliefs, women's role, men's role and their relationship to each other. So the results might be because of the patriarchal society in which we live.

As far as the role of experimental conditions is concerned, the results confirm the findings by Moore (1998) who found a significant effect of the situation (provocative vs non provocative) in which the incident of violence occurs on the attribution of blame of the participants. Findings revealed that in physical violence scenarios, participants labeled the incident as violent but in nonphysical violence scenarios, they did not label it as violent. These findings are consistent with the research findings by Luo (1999). In our culture, people do not perceive nonphysical violence as violence as it is a part of our daily life, and has become the values and norms of our society. It may be because the word 'violence' itself is a very strong word and people only take the physical violence as violence as it has a strong manifestation. In our culture, nonphysical violence is considered as a part of the routine life and people consider it as simple day to day argument between spouses. It is suggested that the definition of violence should be narrowed down and the word should be changed as in the present form it is very broad and people only perceive one form of spousal violence as violence.

Experimental studies usually have problems regarding external validity of its finding. This might be true for this study also, but instead of studying a representative sample, a representative sample of situations were included in the study as participants fulfilling all the criteria were included in the study to control the effect of confounding and extraneous variables making it a factorial design which brings it closer to real life settings. Therefore, we can externally validate the findings with much more confidence than a completely closed setup experiment manipulating a single factor.

With a much larger sample size than the most experimental attribution research, this study is privileged with some degree of representativeness in its findings. Nevertheless, this advantage is not without cost: the findings may be somewhat compromised by the inconsistent cell number for each experimental design. Future studies need to develop better strategies to deal with the dilemma of experimental design vs. representative sample.

The findings of the present study may not be generalizable to a variety of populations, but to make the findings externally valid, it is suggested that similar experiments be conducted using different populations, rural as well as urban, to explore their gender role attitudes and its effect on their attribution of blame for spousal violence. The results of the present study will help learri the mediating role of our gender role attitudes in attribution of blame for spousal violence and it will be a step towards changing the researchers' attitudes towards conducting experiments and may try to remove hesitations and inhibitions regarding the use of experimental method.

Conclusion

patriarchal society like ours, neither men nor women have more modem gender role attitudes. Violence against women is endemic, and is closely linked to society's prescribed gender roles. This may be the reason that majority of the participants held the victim blameworthy instead of the abuser in a spousal violence situation. Further research is needed to fully understand the intricacies of the relationships between the victim, the abuser, and their society. If we really want to minimize the occurrence of role of the moderator and gender role attitudes is one of them.

References

Barbara, A. (2003) the social psychology of minorities. (Report No.38) London: Minority Rights Group. Retrieved March 02, 2004, from vivisome.com.

Ganley, A. L. (1982). Domestic violence: Issues in designing and implementing programs for male batterers. Paper Presented at the American Psychological Association, August, 1978. Retrieved March 2, 2004 from findartoicle.com.

Ho, C. K. (1990). An analysis of domestic violence in Asian American communities: A multicultural approach to counseling. Women & Therapy, 9 (1-2), 129-150. Retrieved March 12, 2004 from findarticles.com.

Kamal, A., & Saqib, T. (2004). Exploring change in gender role attitudes. Paper presented at the National Conference on the Role of Psychology in Socio-economic Development of Pakistan. National Institute of Psychology, Quaid-i-Azam University Islamabad, Pakistan.

Kelley, H. H. (1972). The processes of causal attribution. American Psychology, 28, 107-128. Kenig, H. A., & Ryan, F. (1986). Correlates of life satisfaction among military wives. Journal of Psychology, 123 (5), 465-475. Retrieved January 23, 2004 from looksmart.com.

Locke, L. M. & Richman, C. L. (1999). Attitudes toward domestic violence: race and gender issues. Sex Roles, February. Retrieved January 25, 2004, from www.findarticles.com.

Naseer, S. (2000). Marital adjustment and stress among traditional couples and dual-career couples. (Unpublished M. Phil Thesis), National Institute of Psychology, Quaid-i-Azam University, Islamabad, Pakistan.

Salik, A. (2003). Gender role attitude and its relationship with performance on gender-typed task in performance on gender- typed task in competitive conditions. (Unpublished M.Phil Dissertation). National Institute of Psychology, Quaid-i-Azam University, Islamabad, Pakistan.

Stormo, K. J., Lang, A. R., & Stritzke, W. G. K. (1997). Attributions about acquaintance rape: The role of alcohol and individual differences. Journal of Applied Psychology, 27, 279-305.

Tang, C. S. & Tam, S. Y. (2001). How do Chinese public service professional trainees attribute responsibility to victims and perpetrators of violence against women. Sex Roles: A Journal of Research, April, Retrieved March 12, 2004 from findarticles.com.

Tang, S. K. & Yan, S. (2003). Social construction of women as legitimate victims of violence in Chinese societies. Violence Against Women, 8, 968-996. Retrieved March 19, 2004 from www.findfast.com/spousla violence/articles.html.

Ward, C. A. (1995). Attitudes toward Rape: Feminist and social psychological perspectives. Sage, London.