Rethinking the significance of attitudes in preventing men's violence against women.

Pease, Bob ; Flood, Michael

Introduction

Attitudes have been of central concern in relation to men's violence against women. Community attitudes are seen to play a role in community responses to violence against women, in the perpetration of this violence, and in victims' responses to victimisation. Attitudes have thus been a key target of community education campaigns aimed at preventing violence against women. However, there has been relatively little critical examination of the concept of attitudes and its usefulness in understanding the factors which cause men's violence against women. This article provides such an examination.

The authors of this article were commissioned by a health promotion foundation in Melbourne to review the factors influencing community attitudes in relation to men's violence against women. This literature review was part of a larger project which included a survey of community attitudes on violence against women and an examination of existing campaigns to address community attitudes (Flood and Pease 2006). The public health agency's work was focused on primary prevention in relation to violence against women and changing community attitudes was regarded as a key component of that work. However, as we undertook the literature review, we started to develop some questions about giving primacy to attitudes in developing strategies to address men's violence. In this article, we examine the limitations of prioritising attitudes in violence prevention campaigns.

Attitudes are seen to be significant in shaping violence against women in three key domains: (1) the perpetration of violence against women; (2) women's risk of subjection to violence and response to this victimisation and (3) community and institutional responses to violence against women.

Attitudes and the Perpetration of Violence Against Women

To what extent do attitudes have a causal relationship to the perpetration of violence against women? There is certainly consistent evidence of an association between violence-supportive beliefs and values and the perpetration of violent behaviour, at both individual and community levels. For example, men are more likely to sexually assault if they have hostile and negative sexual attitudes towards women and identify with traditional images of masculinity and male privilege (Heise 1998). Men with more traditional, rigid and misogynistic gender-role attitudes are more likely to practise marital violence (O'Neil & Harway 1997). In a recent meta-analysis aggregating data across all studies relating an aspect of masculine ideology to the incidence of sexual aggression, Murnen et al. (2002) found that all but one measure of masculine ideology were significantly associated with sexual aggression. At the community level, rates of violence against women are higher in contexts where there is widespread acceptance of violence-supportive norms (Heise 1998). A wide variety of studies have found a consistent relationship between men's adherence to sexist, patriarchal, and/or sexually hostile beliefs and their use of violence against women. However, recognition of the role of attitudes in violence against women is only one aspect of a broader, feminist and socio-cultural understanding of this violence. We address these wider issues in a later section of this article.

Attitudes and Subjection to Violence

In relation to the second domain, victimisation, women's responses to their own subjection to violence are shaped by their own attitudes and those of others around them. To the extent that individual women agree with violence-supportive understandings of domestic violence or sexual assault, they are more likely to blame themselves for the assault, less likely to report it to the police or other authorities, and more likely to experience long-term negative psychological and emotional effects. Harris et al. (2005) have demonstrated that women are less likely to report violence and abuse by their partners if they express traditional gender role attitudes. Media portrayals and social norms teach women to 'self-silence', to place their partners' needs above their own (Margolis 1998). Furthermore, stereotypical and narrow representations of violence inhibit women from even recognising and naming their experience as violence. One of the key reasons why women do not report incidents that meet the legal definition of sexual assault is that these do not fit common stereotypes of 'real rape'--they were not by a stranger, did not take place outside and with a weapon, and did not involve injuries (Levore 2003).

Victims also do not report violence because of their perception of others' attitudes: their fear that they will be blamed by family and friends, stigmatised, and the criminal justice system will not provide redress (Levore 2003). In turn, as Koss and Harvey (1991) document in relation to rape, there is an interrelationship between community attitudes, community services, and the psychological experience of the woman raped. While there is evidence that women's recovery from violence is influenced by their attitudes, there is no evidence that attitudes play a causal role in women's risks of victimisation in the first place, and to emphasise this would be to blame the victim for her victimisation. In short, there is no evidence that women's attitudes to rape influence their likelihood of being raped.

Attitudes and Community Responses to Violence Against Women

Attitudes play a role too in the responses and relationships to violence against women adopted by individuals other than the perpetrator or victim. The existence of violence-supportive beliefs means that family members, friends, acquaintances, and bystanders respond with less empathy and support to victims of violence. For example, people who make negative attributions of victims also are less likely to say that they would report the incident to the police and more likely to recommend lenient or no penalties for the offender (Pavlou and Knowles 2001).

Social norms also shape the formal responses of professionals to the victims and perpetrators of violence against women, including police officers, judges, priests, social workers, doctors, and so on. Cross-national studies find that attitudes towards rape and other forms of violence against women inhibit effective and appropriate responses to female victims (Nayak et al. 2003). Among health professionals, those who have received education on child, spouse, and/or elder abuse are more likely than other clinicians to suspect abuse among their clients and to intervene in violence (Tilden et al. 1994). These formal and informal responses have effects on the victims themselves, in that abused women's psychological wellbeing and their ability to escape from abuse are shaped by the levels of material and emotional support they receive (Goodkind et al. 2003).

However, community norms also have a wider significance in sustaining, or sanctioning, violence against women. Because the community is the context in which violence against women occurs, if community norms do not sanction it, they will provide legitimation and support for its continuance. Thus while individual attitudes can change, to achieve widespread change in the attitudes of individuals, it is more important to change community norms (Salazar, 2005). Bem (2002) argues that the most effective way to change 'hearts and minds' is to change behaviour first by changing public policies and practices and by changing social norms. Given that we are all located in social groups which have explicit beliefs, when we stray from those norms, we risk social disapproval.

As community attitudes play a wider role in perpetuating or preventing violence against women, there is a sense in which communities have a collective responsibility for its prevention or perpetuation. Adherence to violence-supportive beliefs by those who may not commit violence creates the atmosphere that encourages other individuals to engage in violence, and non-perpetrating members of groups who perpetrate violence can be seen to be 'morally tainted' by that violence (Radzik 2005). Violence against women can be conceptualised as a hate crime whose attitudinal underpinnings go beyond the perpetrators of such violence to wider community norms (Isaacs 2005). Because the community has some responsibility for the presence and perpetuation of attitudes, so the community has responsibility to challenge culturally pervasive violence-supportive attitudes. For example, violent men may draw on a sense of male entitlement to control women in authorising and legitimating their use of violence (Gilgun and McLeaod 1999). The role of the community then goes beyond dealing with perpetrators to taking on responsibility to challenge the community norms that enable perpetrators of violence to feel comfortable about their beliefs in relation to women. However, as Michalski (2004) asks, to what extent can patriarchal beliefs and attitudes be successfully challenged without changing the social structure within which violence is embedded? We return to this issue later. For now, we are concerned with the construct of 'attitude' itself which underlies community education campaigns against men's violence.

Interrogating the Concept of Attitudes

The construct of attitude is located in the disciplinary field of social psychology. Attitudes have been a central component of social psychology since its beginning and in fact, the discipline of social psychology has been defined as the scientific study of attitudes (Ajzen et al. 2005). The study of attitudes involves an investigation of the factors influencing how they are formed and changed and how they are translated into motivation and behaviour (Albarracin et a., 2005).

There are a variety of different theoretical frameworks for understanding attitudes, and no one theory dominates (Ajzen and Fishbein 2005). In spite of an enormous amount of literature on attitudes, there is no agreed upon meaning for the term (Fabrigar et al. 2005). Nevertheless, several features of attitudes receive consistent emphasis. First, most definitions focus on the process of evaluating an object on a scale ranging from positive to negative.. Eagly and Chaiken (1993), for example, define an attitude as 'a psychological tendency that is expressed by evaluating a particular entity with some degree of favour of disfavour' (cited in Albarracin et al.). Second, attitudes are distinguished from beliefs. It is argued that beliefs can usually be verified or falsified by objective criteria, whereas attitudes cannot be assessed as either true or false (Eagly and Chaiken 1993). Beliefs are said to be based upon knowledge and those holding beliefs tend to believe that such knowledge is correct (Wyer and Albarracin 2005). So while beliefs may be changed by the presentation of factual information, attitudes may be more difficult to change.

Third, social psychological scholarship on attitudes typically views attitudes as stable dispositions. This is in fact regarded as one of their defining features. Most literature assumes that an attitude is accessed from memory and represents a global assessment of the object under scrutiny. In this view attitudes are seen as 'learned structures that reside in the long-term memory and are activated when the issue or object of the attitude is encountered' (Kruglanski and Strobe 2005). In this view, attitudes are regarded as 'enduring psychological constructs that exercise a guiding function on thought and behaviour' (Bassili and Brown 2005: 545). For example, 'sex role attitudes' held by individuals are seen as stable and internally consistent (Billig et al., 1988).

One of the main theoretical notions in attitude research is the idea of the function that the attitude has for the individual and the vested interests that are served by maintaining the attitude. It is argued that attitudes influence behaviour to the extent that they serve the vested interests of the individual (Potter, 1997). This functional approach to attitudes suggests that people hold and express attitudes because of the psychological benefits they derive from them (Herek 1999).

The last two-and-a-half decades has seen the steady development of scholarly tools with which to assess attitudes towards violence against women. Burt's (1980) outline of four key rape myths focused on the victim--nobody was harmed, nothing happened, she wanted or liked it, and she deserved it--was one of the first to operationalise feminist accounts of socio-cultural supports for rape. Two decades later, at least 11 measures of beliefs and attitudes regarding sexual aggression had developed (Murnen et al. 2002), addressing such dimensions of sexual violence as the acceptance of rape myths or adversarial sexual beliefs, hostile or hyper-masculinity, victim-blaming or victim empathy, likelihood of committing rape if one was assured of not being caught, and actual sexually coercive behaviour. Other instruments focus on attitudes towards and perceptions of other, specific forms of violence against women, from wife assault to sexual harassment and date rape. Many attitudinal instruments use Likert scales in which respondents 'agree' or 'disagree' with a series of statements, some supportive of violence against women and some not.

While this research on attitudes to violence has expanded in the last two decades, traditional psychological accounts of attitudes have been subject to growing critique within the social sciences during the same time frame. First, there is evidence that attitudes are contextual and contingent. While most attitude researchers assume that attitudes reside within people and are enduring, some research suggests that attitudes are often simply 'temporary constructions created at the time people are asked to make attitudinal judgements' (Fabrigar et al. 2005: 80). Just because an attitude is expressed on one occasion, does not mean that that the same attitude will necessarily be expressed on another occasion. Potter (1998) maintains that individuals will sometimes offer different attitudes even during the same conversation. It has been suggested that people's mood at the time of the interview will be significant in shaping their attitudinal responses to a phenomena. People in a positive mood tend to evaluate events more favourably than people in a negative mood (Ajzen and Fishbein 2005). Potter (1997) argues that such variability is often suppressed because it contradicts those theories which explain behaviour on the basis of underlying and consistent evaluative positions. In short, attitudes are not necessarily stable, easily quantifiable or unitary (Tuffin, 2005).

To what extent do expressed attitudes allow us to predict behaviour? There is growing evidence that the relationship between attitudes and behaviour is both complex and partial. For example, human behaviour is not necessarily reasoned or planned, and the influence on behaviour of rational and cognitive processes may be only as important as emotions and non-cognitive influences (Ajzen and Fishbein, 2005). There is considerable research to show that people might say one thing and do another. Ajzen and Fishbein (2005: 175) cite research that demonstrates that attitudes are 'very poor predictors of actual behaviour'. In the area of prejudice and discrimination, Fiske (1998) has demonstrated that while 'expressions of stereotypical beliefs and prejudicial attitudes have declined markedly over the past decades, discrimination against historically disadvantaged racial and ethnic groups continues to be evident in employment, education, housing, health care and criminal justice' (cited in Ajzen and Fishbein 2005: 204). So prejudice has declined considerably more than discrimination. Community attitudes may show greater intolerance towards violence but this does not necessarily mean that violence will decline as a result.

In an attempt to explain this disparity, some researchers distinguish between implicit and explicit attitudes. Basili and Brown (2005: 546) argue that 'implicit attitudes represent a more accurate reflection of people's inner feelings than explicit attitudes'. In their view, implicit attitudes are more influential is shaping how we think and act. This notion is offered as one explanation to explain the gap between high levels of discrimination against women and apparent change in stereotypical attitudes. So people can hold explicit egalitarian attitudes and at the same time hold implicit prejudiced attitudes. Basili and Brown (2005) point out that standard attitude scales are unable to measure implicit aspects of prejudice.

This recognition has implications too for the ways in which we measure attitudes. For example, quantitative surveys are less adept at tapping cultural scripts and implicit cultural norms than more qualitative methods. As Vandello and Cohen (2003) note, 'there is likely to be an important disjunction between consciously articulated, explicit condemnation of domestic violence and a more implicit approval of the scripts, norms, and roles that lead to such violence'. Their own study confirmed this, in that an experimental method picked up cultural differences that the attitudinal items did not.

If attitudes do not have a simple and determining impact on behaviour, there is good reason to be cautious about focusing on attitudes as the most important target of violence prevention efforts. Community education campaigns are a common strategy of violence prevention. While some campaigns have been shown to produce substantial and lasting change in community attitudes, many have not, and changes in attitudes may not result in changes in behaviour (authors, 2006). Sless (1998) raises questions about the value of both studies that set out to measure attitudes and campaigns designed to try and change attitudes, arguing that we are more likely to change people's behaviour by changing the environment in which they act and the sanctions or permissions that govern their actions. Shrensky (1998) also criticises traditional attitude theory which posits that attitudes cause behaviour, arguing that behaviour needs to be understood within particular social and cultural contexts.

Much attitude theory draws for its understanding upon the intra-psychic world emphasised in cognitive psychology, in which attitudes are seen as underlying mental constructs. Critical psychologists, using an epistemology emphasising the social, challenge the notion that attitudes reside within the individual psyche (Tuffin 2005). Critical research in this area is concerned with social and ideological conflicts and practices in particular settings. For example, Wetherell and Potter (1992) in their study of racism in New Zealand moved away from individual attitudes held by participants to focus on how they dealt with ideological dilemmas. This involves a shift from the notion that people carry Likert scales in their heads towards a much more complex understanding of how evaluative practices are enacted in different settings.

The assumption underpinning most social influence theorising is that each individual 'constructs a sense of self that is separate and independent from others' (Prislin and Wood 2005: 693). Potter and Wetherell (1987) point out that people's behaviour is only partly influenced by their attitudes. People's actions are also shaped by norms and what other people will think of their behaviour. So personal dispositions may be less important than the social context and social norms in determining behaviour.

Social norms are defined by Prislin and Wood (2005: 677) as 'shared belief systems about what people do or what they ideally should do'. The more an individual accepts the attitudes of others, the greater the importance of changing the social context in influencing that individual's views. Herek (1999: 14) has demonstrated the implications of a social constructionist view of attitudes in relation to attitudes about AIDS by showing the importance of 'grounding an attitude domain within a specific social group'.

What this means is that attitudes can be shaped by a new social consensus. If we change the social context, then we can change attitudes. Prislin and Wood (2005: 672) demonstrate how 'changes in people's social environments especially the pattern and content of their social interactions effect changes in social attitudes'. In this view, attitudes are socially constructed and they change when the social context changes. This involves a shift from understanding attitudes as individual processes to seeing them in the context of social relations. The opinions held by others are thus significant in shaping individuals' responses to particular phenomena. Individuals' responses and attitudes towards an issue are influenced by what they understand the consensus to be on that issue, by their perceptions of dominant norms (Prislin and Wood 2005). If attitudes are socially constructed and based on shared knowledge that people use to make sense of the world, this means that the concepts of 'social norms' and 'dominant ideologies' in relation to men's violence against women may be more useful than the psychological construct of 'attitudes' in addressing this issue.

We acknowledge that many psychologists would maintain that norms are simply a reflection of 'societal attitudes'. Danziger (1997), for example, notes the distinction between the concept of 'social attitudes' and that of 'ideology'. Whereas the concept of ideology has moral connotations and suggests an imperative towards social change, the language of societal attitudes has a morally neutral tone. Also while ideology is a socio-political framework that focuses on individuals' membership of social groups, societal attitudes are regarded as individual attributes. From a psychological perspective, attitudes are only social because they are seen as individual responses to social stimuli. Thus social phenomena like collective values are reduced to the behaviour of individuals. An understanding of ideology is more concerned with how individual character is shaped by social conditions and social consciousness.

Thus far, we have argued that individuals' evaluations and understandings of men's violence against women, their 'attitudes' towards the phenomenon, must be recognised as contingent, contextual, potentially contradictory, having a complex relationship to behaviour, and constructed and meaningful only in social contexts. It would be premature at this stage to abandon the notion of 'attitudes' in assessing and addressing violence against women. At the same time, the critical and social constructionist understanding of attitudes offered here will enhance the practical utility of the concept's deployment and the theoretical insight this generates. We move now to a wider examination of the significance of attitudes per se in understanding and challenging men's violence against women.

Moving Beyond Attitudes in Challenging Men's Violence Against Women

While attitudes towards violence against women are important influences on the perpetration of violence against women, women's responses to this victimisation, and wider community and institutional responses, we need to locate this in a wider context and note the key extra-attitudinal influences on violence against women. Having established that attitudes are complex and contextual, we must also recognise that they do not sufficiently explain violence against women, and therefore they should not be the only target of prevention efforts.

In explaining violence against women, attitudes should not be taken to be the only variable of concern, nor even the primary one. In the first place, a single-minded focus on attitudes risks over-emphasising the cognitive elements of individual perceptions of violence against women. Similarly, cognitive motivations are only one aspect of an explanation of individuals' perpetration of violence against women. When a man physically assaults his wife or sexually harasses his female colleague, this behaviour may be shaped in part by his adherence to violence-supportive attitudes, but it may also be shaped by his affective orientations and other aspects of his identity or subjectivity. For example, in a study among university undergraduates, Hill and Fisher (2001) found that men's sense of entitlement, both general and specifically sexual, mediated the relationship between masculine gender roles and sexually aggressive behaviour and attitudes. 'Entitlement' refers here to men feeling entitled to have their needs met by women and believing that their needs or desires take precedence over women's. While there was little direct relationship between men's attitudes towards male gender norms and various rape-related variables (their acceptance of rape myths, likelihood of committing rape, victim-blaming, and actual sexually coercive behaviour), masculinity factors predicted men's general and sexual entitlement, and these in turn predicted an array of rape-related attitudes and behaviours.

Men's use of violence against women clearly is shaped by patriarchal, anti-women, and 'hypermasculine' ideologies, as Murnen et al's (2002) meta-analysis documents in relation to sexual aggression. However, the possession of such attitudes is not necessarily sufficient to perpetrate sexually aggressive behaviour. Instead, such attitudes combine with situational factors to predict violence against women, such as the presence of situations in which coercive sexual encounters can occur, heavy alcohol consumption (which men may use to minimise their responsibility), peer pressure and peer support for perpetration, and so on (Murnen et al. 2002). Indeed, adherence to violence-supportive attitudes in some instances may not even be a necessary condition of violence perpetration. In a study among American undergraduates, Locke and Mahalik (2005) found that some men did not endorse rape myths but did report sexually aggressive behaviours. They hypothesise that these men may become sexually aggressive after drinking or engaging in risky behaviours. Attitudes which are intolerant of violence may be only weakly held, and may be 'rendered situationally inoperative' or neutralised by situational variables (Sellers et al. 2005).

Continuing to focus on sexual violence, different factors may be involved in different forms of sexual aggression or for different types of sexually aggressive men, with attitudes playing lesser or greater roles depending on these. Some perpetrators may be highly aroused by sexual violence and likely to commit multiple acts of aggression with different victims, while others may be more influenced by cognitive motivations and more likely to commit assaults in situations where they perceive or can argue for some justification for their behaviour (Murnen et al. 2002).

Nor should we assume that violence against women is shaped above all by attitudes. A focus on individual attitudes risks neglecting the cultural, collective, and institutional underpinnings of violence against women. Beyond individual attitudes and perceptions, violence against women is shaped by the social, cultural, economic, and political relations of particular contexts, communities, and cultures. While these collective social relations do have attitudinal dimensions, they are not reducible to them.

There is good evidence that violence against women is shaped by extra-attitudinal features of formal and informal contexts, networks, and institutions--of families, workplaces, neighbourhoods, social networks, and peer groups. For example, at the level of the immediate context in which violence against women takes place--typically families or other intimate or acquaintance relationships--patterns of male dominance are associated with higher rates of violence. Cross-culturally, male economic and decision-making dominance in the family is one of the strongest predictors of societies showing high levels of violence against women. Wife abuse is more likely in couples with a clearly dominant husband, while a wife's economic dependence on her husband is a major predictor of severe wife beating and of marital rape (Heise, 1998). Michalski (2004) concurs, noting the evidence that centralised patriarchal authority in families predicts higher rates of violence. Domestic violence is shaped also by the frequency and intensity of marital conflicts, and when such conflict occurs in families with an asymmetrical power structure, there is a much higher risk of violence (Heise 1998).

Further contextual factors shape violence against women. Although violence against women occurs in all classes, there is strong evidence that wife abuse is more common in families and communities characterised by low socioeconomic status and or unemployment (Heise 1998). Low socioeconomic status may increase the risk of abuse because of the other variables which accompany this, such as crowding, hopelessness, stress, or a sense of inadequacy in some men. Poverty may provide fodder for marital disagreements, or make it harder for women to leave violent or otherwise unsatisfactory relationships. To take another neighbourhood-level factor, social isolation is both a cause and a consequence of wife abuse. Women with strong family and friendship networks experience lower rates of violence (Heise 1998). Michalski (2004) makes the same point, that integrated social networks and cross-cutting intimate ties curtail the likelihood of violence occurring during disputes. A third factor is peer associations. Especially among adolescent males, peer group behaviours play an important role in encouraging sexual aggression. While we noted the attitudinal dimensions of involvement in violence-supportive peer contexts earlier in this review, other dimensions include participation in group leisure activities, 'partying', and so on. For example, men's workplace 'girl watching' (often a type of sexual harassment) can function as a form of gendered play among men, in which men objectify women as part of 'games' and contests for status with each other (Quinn 2002).

We thus argue that attitudes should be grounded within specific social settings and among specific social groups. As we have argued earlier, people's behaviour is shaped by social norms and what other people think of their behaviour. 'Attitudes' are thus shaped by the social consensus within specific settings and among specific groups. Particular institutions such as schools, workplaces, or churches shape their participants' attitudes through both formal policies and structures and informal norms. They are locations for informal peer relations which shape attitudes, and such institutions are themselves shaped in dynamic ways by wider factors such as the mass media. Consequently, to change the attitudes of individuals, we need to challenge the dominant norms within those settings and among those groups in which those individuals are situated. Violence against women is also constituted by wider historical and political forces and contexts, whether histories of colonisation and the disintegration of family and community in indigenous communities or wars, civil political conflicts, militarism and imperialism, and other global forces (Greig et al. 2000).

The discussion thus far has focused on extra-attitudinal influences, including contextual and situational factors, on the perpetration of violence against women, but such factors also shape formal and informal responses to this violence. For example, family and friends' reactions to a battered woman depend on contextual factors such as the woman's relationship with her assailant, the number of children, and whether family and friends themselves are threatened with violence (Goodkind et al. 2003).

We have argued that explanations of violence against women must be grounded not primarily in individually-held attitudes but in social and cultural norms and ideologies. However, we acknowledge that when we act in the world, we are not just operating within structural constraints. We are also determining the nature of those structures through our actions and interactions. This means that we can challenge those arrangements. Our argument here, however, is that the structural dimensions of men's violence against women cannot be transformed by simply changing individual and community attitudes alone. Explanations of this violence, and thus interventions to prevent it, must also be grounded in social relations and social structures.

We must move beyond a strictly cultural emphasis in both explanation and intervention, recognising that 'violence has much deeper roots in the structural foundations of interpersonal relationships (and societal arrangements in general)' (Michalski, 2004: 653). Authors such as Heise (1997) and Michalski (2004) identify key features of collective social relations which help to explain the persistence and pervasiveness of violence against women, as we have noted above. Thus, to stop violence against women, we must do more than change community attitudes. We must also address 'the structural conditions that perpetuate violence at the interpersonal and even societal level' (Michalski 2004: 670).

Implications for community education campaigns to change community attitudes

Given the breadth of factors, settings, and social forces which shape men's violence against women, we need to locate intervention strategies in a wide range of possible settings and groups. The process of 'changing attitudes' must be located within a project of changing familial, organisational, community and societal norms which support violence against women. Interventions must address not only those attitudes which are overtly condoning of violence against women, but the wider social norms related to gender and sexuality which normalise and justify this violence. Given the close association between attitudes towards violence against women and beliefs about gender roles, especially males' adherence to sexist, patriarchal, and/or hostile views towards women, traditional gender roles must be targeted in intervention campaigns. Efforts to address violence-supportive norms must work also to provide an alternative, a set of violence-intolerant norms and values centred on non-violence, gender equality, and social justice. Finally, interventions aimed at cultural change must be accompanied by changes in structural relations and social practices if violence against women is to be undermined and prevented.

References

Ajzen, I. and Fishbein, M. (200,5) 'The influence of attitudes on behavior'. In D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.

Albarracin, D., Johnson, B., Zanner, M. and Kumkale, G. (2005) 'Attitudes: Introduction and scope'. In D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.

Bassili, J. and Brown, R. (2005) 'Implicit and explicit attitudes: Research, challenges and theory'. In D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London:, Lawrence Erlbaum Associates.

Bem, D. (2002) 'Introduction to Beliefs, Attitudes and Ideologies'. Unpublished paper.

Billig, M., Condor, S., Edwards, D., Gane, M., Middleton, D. and Radley, A. (1988) Ideological dilemmas: A social psychology of everyday thinking, London, Sage.

Burt, M. (2003) 'Cultural myths and support for rape', Journal of Personality and Social Psychology, 38, 217-230.

Danziger, K. (1997) Naming the Mind: How Psychology Found its Language, London, Sage.

Eagly, A. and Chaiken, S. (1993) The Psychology of Attitudes, Orlando FL, Harcourt Brace Jovanovich.

Fabrigar, L., Macdonald, T. and Webener, D. (2005) 'The structure of attitudes', In D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.

Fiske, S. (1998) 'Sterotyping, prejudice and discrimination'. In D. Gilbert, S. Fiske and L. Gardner (eds) The Handbook of Social Psychology, Vol. 2, 4th Ed, Boston, MA: McGraw-Hill.

Flood, M. and Pease, B. (2006) The Factors Influencing Community Attitudes in Relation to Violence Against Women: A Critical Review of the Literature, Melbourne, Victorian Health Promotion Foundation

Gilgun, J., and L. McLeaod (1999) 'Gendering violence', Studies in Symbolic Interaction, 22, 167-193.

Goodkind, J.R., T.L. Gillum, D.I. Bybee, and C.M. Sullivan (2003) 'The impact of

family and friends' reactions on the well-being of women with abusive partners', Violence Against Women, 9 (3), 347-373.

Greig, A., and D. Peacock (2005) Men as partners programme: Promising practices guide. Engender Health (Draft only).

Harris, R., J. Firestone and W. Vega (2005) 'The interaction of country of origin, acculturation and gender role ideology on wife abuse', Social Science Quarterly, 86 (2), 465-483.

Heise, L. (1998) 'Violence against women: An integrated, ecological framework', Violence Against Women, 4 (3), June, 262-283.

Herek, G. (1999) 'The social construction of attitudes: Functional consensus and divergence in the US public's reaction to AIDS'. In G. Maio and J. Olson (Eds.) Why We Evaluate: Functions of Attitudes, Mahwah, NJ, Lawrence Erlbaum.

Hill, M.S., and A.R. Fischer. (2001) 'Does entitlement mediate the link between masculinity and rape-related variables?'. Journal of Counseling Psychology, 48 (1), 39-50. Koss, M.P., and M.R. Harvey (1991) The Rape Victim: Clinical and Community Interventions (2nd ed.) Newbury Park, CA., Sage.

Kruglanski, A. and Strobe, W. (2005) 'The influence of beliefs and goals on attitudes: Issues of structure, function and dynamics'. In D. Albarracin, B. Johnson and M. Zanna (eds) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.

Levore, D. (2003) Non-reporting and Hidden Reporting of Sexual Assault: An international literature review. Canberra: Australian Institute of Criminology.

Locke, L. and Mahalik, J. (2005) 'Examining masculinity norms, problem drinking and athletic involvement as predicators of sexual aggression in college men', Journal of Counseling Psychology, 52 (3), 279-283.

Margolis, D. (1998) Culturally Sanctioned Violence Against Women: A Look at Attitudes Toward Rape. Unpublished PhD dissertation, Graduate School of Education, Boston College, Boston.

Michalski, J. (2004) 'Making sociological sense out of trends in intimate partner violence: The social structure of violence against women', Violence Against Women, 10 (6), 652-675.

Murnen, S.K., C. Wright, and G. Kaluzny (2002) 'If 'boys will be boys,' then girls will be victims? A meta-analytic review of the research that relates masculine ideology to sexual aggression', Sex Roles, 46 (11-12), 359-375.

Nayak, M., C. Byrne, M. Martin, and A. Abraham (2003) 'Attitudes towards violence against women: A cross-nation study', Sex Roles, 49 (7/8), 333-342.

O'Neil, J.M., and M. Harway. (1997) 'A multivariate model explaining men's violence toward women: Predisposing and triggering hypotheses', Violence Against Women. 3 (2), 182-203.

Pavlou, M., and A. Knowles (2001) 'Domestic violence: Attributions, recommended punishments and reporting behaviour related to provocation by the victim', Psychiatry, Psychology and Law, 8 (1), 76-85.

Potter, J. (1998) 'Discursive social psychology: From attitudes to evaluations', European Review of Social Psychology, 9, 233-266.

Potter, J. and Wetherell, M. (1987) Discourse and Social Psychology: Beyond Attitudes and Behaviour, Newbury Park CA, Sage.

Prislin, R. and Wood, W. (2005) 'Social influence in attitudes and attitude change'. In

D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.

Quinn, B.A. (2002). 'Sexual harassment and masculinity: The power and meaning of "girl watching"', Gender & Society, 16(3), 386-402.

Radzik, L. (2005) 'Collective responsibility and duties to respond', Social Theory and Practice, 27 (3), 1-16.

Salazar, L. (2005) 'Moving beyond the individual: Examining the effects of domestic violence policies on social norms', American Journal of Community Psychology, 32 (3-4), 1-12.

Sellers, C., J. Cochran, and K. Branch (2005) 'Social learning theory and partner violence: A research note', Deviant Behavior, 26 (4), 379-395.

Tuffin, K. (2005) Understanding Critical Psychology, Newbury Park, CA, Sage.

Shrensky, R. (1998) The Ontology of Communication. Unpublished PhD Thesis, University of Canberra, Canberra.

Sless, D. (1998) The problem with attitude. Melbourne: Communication Research Institute.

Tilden, V., T. Schmidt, B. Limandri, G. Chiodo, M. Garland, and P. Loveless (1994) 'Factors that influence clinicians' assessment and management of family violence', American Journal of Public Health, 84 (4).

Vandello, J.A., and D. Cohen (2003) 'Male honor and female fidelity: Implicit cultural scripts that perpetuate domestic violence', Journal of Personality and Social Psychology, May, 84(5), 997-1010.

Wetherell, M. and Potter, J. (1992) Mapping the Language of Racism: Discourse and the Legitimation of Exploitation, London, Harvester.

Wyer, R. and Albarracin, D. (2005) 'Belief formation, organisaton and change: Cognitive and motivational influences'. In D. Albarracin, B. Johnson and M. Zanna (Eds.) The Handbook of Attitudes, London, Lawrence Erlbaum Associates.