• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of jepicomhJournal of Epidemiology and Community HealthVisit this articleSubmit a manuscriptReceive email alertsContact usBMJ
J Epidemiol Community Health. Feb 2006; 60(2): 123–129.
PMCID: PMC2588066

Acceptability of domestic violence against women in the European Union: a multilevel analysis

Abstract

Study objective

The acceptability of domestic violence against women (DVAW) plays an important part in shaping the social environment in which the victims are embedded, which in turn may contribute either to perpetuate or to reduce the levels of DVAW in our societies. This study analyses correlates of the acceptability of DVAW in the European Union (EU).

Design

Three level ordinal logistic regression of 13 457 people nested within 212 localities (cities), nested within 15 countries of the EU. Sampling is multistage with random probability. All interviews were face to face in people's homes. The outcome variable was acceptability of DVAW. Multiple correlates at the individual, locality, and country level were analysed.

Setting

European Union, 1999.

Participants

National data were used of residents 15 years old and above of all member states in 1999 (n = 13 457). Average response rate was 47%, although it varied across countries (23%–73%).

Main results

Higher levels of acceptability were reported by those who perceived DVAW as less severe and less frequent. Acceptability is higher among men who know a perpetrator and lower among men who know a victim. Victim blaming attitude is associated with higher levels of acceptability. In countries with higher gender empowerment measure the difference in acceptability among those who blame and those who do not blame the victim is greater.

Conclusions

There are still widespread attitudes in the EU such as victim blaming that condone DVAW, contributing to a climate of social acceptability of DVAW. Further efforts to reduce the acceptability of DVAW are still needed.

Keywords: domestic violence, violence against women, attitude, multilevel analysis, European Union

The extent of domestic violence against women (DVAW) worldwide, which in western countries affects about 25% of women at some point in their lives,1,2,3,4,5 its impact on their physical and mental health, both in the short and long term, and the wider outcomes of DVAW for families, communities, and society, makes it a public health priority.6,7,8,9,10

A public health approach to DVAW, as the one proposed by the World Health Organisation,10 considered that among the larger societal factors that influence rates of violence are those that create an acceptable climate for violence, and those that reduce inhibitions against violence. Although the influence of contextual factors in DVAW has been theoretically acknowledged, research examining these effects has been scarce. A small number of multilevel studies showed that social and community level factors such as neighbourhood poverty and disadvantage, are related to higher rates of DVAW.11,12,13,14,15,16 However, to our knowledge, no multilevel study has analysed the individual and social factors associated with public attitudes toward DVAW. Clearly, a better understanding of the factors influencing public attitudes toward DVAW would add relevant knowledge to the literature on the social conditions that foster or discourage DVAW, and would be useful to orientate public education and prevention efforts.

Domestic violence is always rooted in a social and cultural context, and public attitudes about what is or is not acceptable in intimate relationships reflect these social and cultural norms.17,18,19,20,21,22,23 As it has been emphasised by a number of scholars, without a fundamental change in the social attitudes that foster, condone, and perpetuate DVAW we will not be able to respond effectively to this problem, by substantially reducing its alarming rates.17,18,19,24,25

Public attitudes about DVAW play an important part in shaping the social environment in which the victims are embedded, which in turn may contribute either to perpetuate or to reduce the levels of DVAW in our societies.6,17,18,19 Family, neighbours, friends, coworkers, and even acquaintances, are an important part of this social environment that may act as potential guardians whose intervention to help victims of domestic violence may depend on public perceptions of the unacceptability of all or only certain types of incidents.26 If DVAW is considered as such only when it involves extreme, severe, or repeated violence, it is more likely that some violence towards women in intimate relationships may be acceptable under some conditions.27,28 A social environment that accepts or even supports domestic violence in some circumstances contributes to create a climate of tolerance that makes it easier for perpetrators to persist in their violent behaviour, and makes it more difficult for women to disclose domestic violence.6,17,25,29

If primary prevention is a cornerstone of a public health approach, reducing the acceptability of all forms of DVAW becomes one of its fundamental goals. In this effort, public education and media campaigns challenging social attitudes towards DVAW are basic tools to approach social levels of acceptability to zero.6 However, little research attention has been directed to explore public levels of acceptability of DVAW and to examine its correlates in our societies so that primary prevention efforts can be better informed. Prevention policies would indeed benefit from data monitoring the epidemiology of the acceptability of DVAW in our societies to design and evaluate the outcome of public education efforts.17 The aim of this paper is to address this gap in our knowledge by analysing the acceptability of DVAW and its correlates in a representative sample of Europeans of all member states of the European Union in 1999. Using a multilevel approach, individual, locality, and country level correlates of the acceptability of DVAW will be analysed. At the individual level, sociodemographic factors will be explored as they have been found to be related to DVAW.30,31,32,33 The influence of individual perceptions of the frequency and severity of DVAW, knowledge of victims and perpetrators of DVAW, and victim blaming attitudes will also be explored as they may affect levels of acceptability.6,17,27,28 At the locality and country level, we analyse socioeconomic indicators as multilevel studies have shown that socioeconomic characteristics at the aggregate level are related to DVAW.11,13,14 Also, at the country level, gender inequality will be analysed as research has found that low egalitarianism as well as changes in gender norms toward a higher status of women in society are associated with higher rates of DVAW.34,35,36,37

Methods

Data from the Eurobarometer 51.0 “Europeans and their views on domestic violence against women” was used for this study. It was carried out in 1999 and covered the population of residents 15 years and over in each of the member states of the EU at that time.38 Sample design applied in all member states is multistage with random probability. In each EU country, localities (cities) were drawn with probability proportional to population size (for a total coverage of the country) and to population density. In each locality, a starting address was drawn at random and further addresses were selected by standard random route procedures from the initial address (that is, taking alternate left and right hand turns for selecting blocks and apartment buildings and selecting households randomly calling at every Nth household). In each household, the respondent was drawn at random. All interviews were face to face in people's homes and in the appropriate national language. Data are from 13 457 Europeans of 212 localities (cities) of all member states of the EU at the time of the survey (year 1999). Table 11 shows country, response rates for each country, final sample, number of localities sampled per country, and average number of respondents per locality.

Table thumbnail
Table 1 Response rates and final sample by country. Eurobarometer 51.0, European Union, 1999

Although the survey did not provide a general definition of DVAW, it was presented to respondents as comprising five different types of which they were asked their opinion: sexual violence, physical violence, psychological violence, restricted freedom, and threats.

Outcome variable

The outcome variable is acceptability of DVAW, measured with the following question: In your opinion, is domestic violence against women…? Possible responses were: 1, unacceptable in all circumstances and always punishable; 2, unacceptable in all circumstances and not always punishable; 3, acceptable in certain circumstances; and 4, acceptable in all circumstances.

Predictors

Subjects

Sex (1 = male, 0 = female). Knows a victim of DVAW (1 = yes, 0 = no) and knows a perpetrator of DVAW (1 = yes, 0 = no) in respondents' social circle of work, studies, family, neighborhood and friends. Perceived frequency of DVAW (1 = very common or fairly common, 0 = not very common or not at all common). Perceived lack of severity of DVAW evaluates how the respondent considers the five different forms of DVAW (1 = very serious, 2 = fairly serious, 3 = not very serious, and 4  = not at all serious). We summed up scores of every form of DVAW to compute a unique score for each person's perception of lack of severity of DVAW (range 5 to 20). Cronbach's α value for this scale was 0.80. Victim blaming attitude (1 = yes, 0 = no) (see table 33 for a complete description of questions).

Table thumbnail
Table 3 Distribution of DVAW related variables by sex. Eurobarometer 51.0, European Union, 1999

Countries

We used the gender empowerment index (GEM) to capture gender inequality in three key areas: political participation and decision making power, as measured by women's and men's percentage shares of parliamentary seats; economic participation, and decision making power, as measured by women's and men's percentage shares of positions as legislators, senior officials and managers, and women's and men's percentage shares of professional and technical positions; and power over economic resources, as measured by women's and men's estimated earned income. A higher value indicates a higher level of gender empowerment. GEM indices are published by the United Nations Development Programme.39

Because of the model selection strategy implemented in this study (see below, analytical strategy and model selection) variables at the individual, locality, and country level that did not significantly reduce model deviance were not retained in the final model and are not presented in the results section. For the interested reader, these variables were the following:

Individual level: age, education, household income, marital status (married, never married, divorced/separated, and widowed), knowledge of special laws to combat domestic violence in their country (regarding both the victim and the perpetrator of domestic violence), job status (employed, unemployed; white collar, blue collar), main household earner (whether respondent is the person in the household who contributes most to the household income), main household worker (whether respondent is the person in the household mainly responsible for ordinary shopping and looking after the home), political ideology (how would you place your political views from 1 = left to 10 = right?). Locality level: all individual variables aggregated at the locality level. Country level: unemployment, education, rates of divorces and gross national product, retrieved from international databases for each country.

All explanatory variables were centred on their grand mean (for continuous variables) and their grand mode (for categorical variables) to ease interpretation of the parameter estimates.

From the initial 16 179 people surveyed, 4% (n = 490) did not express an opinion to the interviewer about acceptability of DVAW (or other missing values in variables of interest—that is, sex and age of interviewer). Also, as “do not know” and “no answer” responses were hard to interpret, we treated them as missing values and excluded them for the analysis. This led to the final sample reported here (n = 13 457), which equals 85% of respondents. Given this loss of sample, we also recoded variables of the study so that higher values (“1”) reflected high levels of each variable (that is, high perceived frequency) and low values (“0”) reflected either low levels or “do not know” responses, thus increasing final sample size to 15 684 (96% of those initially surveyed). This was the upper limit of sample size that we could reach because missing values on sex and age of interviewer and acceptability of DVAW could no be imputed. Next, we calculated the final model reported in this research and results were virtually the same. Although this strategy increased sample size (n = 15 684), it posed serious problems of interpretation. Mainly, “do not know” responses have no intrinsic meaning and could be an important source of error (that is, disinterested or uncooperative respondent, etc). Because of this, we decided to restrict the final sample size thus gaining in interpretability of results. Below we present results for complete cases (n = 13 457).

Analytical strategy and model selection

Data present a clear multilevel structure with subjects (level 1) nested within localities (level 2) nested within countries (level 3). We used multilevel modelling that permits the inclusion of additional error terms that reflect the complex pattern of variation introduced by the hierarchical structure of the data (random effects).40 Because the scoring system of the outcome variable was arbitrary, information could be lost or distorted in the conversion to a continuous variable.41 Therefore, we treated the four responses as ordered categorical and conducted three level ordinal logistic regression analyses using the HLM3 module of the statistical package HLM 6.01.42

We checked for multicollinearity problems among predictors examining the variance inflation factor (VIF), all off diagonal elements in the variance‐covariance (τ) matrix for correlations close to 1 or −1, and the diagonal elements for any elements close to zero, with no indication of multicollinearity.42

The multilevel analysis was performed in steps. The starting point was an empty model without explanatory variables in which the total variance of acceptability of DVAW was partitioned into a component at each level (model 1). In the next step we explored fixed effects (main as well as interaction) of variables at the individual, locality, and country level (model 2). At this step, we explored possible interactions between all predictor variables. In the last step random slopes between localities and between countries were examined for DVAW related variables and characteristics of interviewer. This final model (model 3) is therefore an extension of model 2 and the only difference between these two models is that model 3 incorporates random effects. Full penalised quasi‐likelihood (PQL) was used to estimate parameters in the model.42 For model selection, models with smaller deviance (likelihood function) were selected.

Results

Table 22 shows descriptive statistics. For perceived lack of severity of DVAW, descriptives are: range 5–20; mean, 6.49; and standard deviation, 1.93. Table 33 presents the distribution of DVAW related variables by sex. Women knew more victims and more perpetrators, perceived more frequency, more severity, and accepted DVAW less than men. Men considered more frequently than women that the provocative behaviour of women is a cause of DVAW. Beyond these sex differences, victim blaming attitudes seem to be widespread in the sample (about 56% of men and 53% of women).

Table thumbnail
Table 2 Frequency of selected sociodemographic, outcome, and predictor variables of the Eurobarometer 51.0, European Union, 1999

Table 44 presents parameter estimates for the three models. Model 1 shows that there is significant variability of acceptability of DVAW across localities (0.26, SE = 0.04) and countries (0.26, SE = 0. 11). Model 2 reduces the variation of intercepts between countries although there is still significant variation between localities (0.26, SE = 0.04) that is not accounted for by the model. This may show that differences in acceptability between localities are hidden by their individual composition.40 In model 3, the inclusion of random effects at level 2 reduces both the variation of intercepts between localities (0.21, SE = 0.03) and between countries (0.19, SE = 0.08). These random effects were: sex and age of interviewer, and perceived frequency of DVAW. Also, deviance is smaller for model 3 than for the other two models (the smaller, the better). The relative reduction of variance from model 1 to 3 is 20.7% in level 2 and 26.0% in level 3, as indicated by PCV (proportional change variance).43

Table thumbnail
Table 4 Estimates, standard errors and 95% confidence intervals for fixed and random effects from three level multilevel analysis of the Eurobarometer 51.0, European Union, 1999†

The intercept (γ000) captures the overall log‐odds of accepting DVAW after controlling for all the predictors at the grand mean level. For easier interpretation, the log‐odds can be transformed to a probability scale as follows: p = exp(γ000)/(1+exp(γ000)). This formula gives the probability of being in the category 1 (acceptable in all circumstances), which is 0.004 (0.4%) in model 3. Threshold δ(2) is the difference in the log‐odds of category 2 versus 1 and threshold δ(3) is the difference in the log‐odds between categories 3 compared with 1 and 2, holding constant the fixed and random effects of the model. Adding the threshold δ(2) (see bottom of table 44)) gives the probability of categories [less-than-or-eq, slant]2 (always accepting DVAW or only in certain circumstances). For model 3: p = exp(–5.29+1.28)/(1 + exp(–5.29+1.28)) = 0.017 (1.7%). Finally, adding threshold δ(3) gives p = 0.354 (35.4%), which is the probability of category [less-than-or-eq, slant] 3. This probability minus 0.017 (p = 0.354–0.017 = 0.337) is the probability of category 3 (unacceptable and not always punishable). Likewise, the probability of category 4 (unacceptable and always punishable) equals 0.646 (1−0.354 = 0.646). This means that respondents had 1.7% probability of accepting DVAW (categories [less-than-or-eq, slant] 2), 33.7% of being in category 3 (unacceptable and not always punishable), and a probability of 64.6% of being in category 4 (unacceptable and always punishable), after adjusting for all the covariates in the study. These percentages sharply contrast with the probability of acceptability for a man who blames the victim, perceived that DVAW is rather uncommon and non‐severe, does not know a victim, and knows an aggressor of DVAW. For this person, the probability of accepting DVAW is 7% (categories 1 and 2), about five times above the average person of the EU. Likewise, his probability of being in category 4 (unacceptable and always punishable) is 21% (about one third of the probability of the average person).

As for the fixed effects, blaming women for DVAW and perceiving lack of severity are positively and significantly associated with acceptability of DVAW, whereas perceived frequency is significantly and negatively associated with acceptability. Sex has an interaction effect with knowing a victim and knowing a perpetrator of DVAW: men who know a victim of DVAW have lower probability of accepting DVAW whereas in men who know a perpetrator this probability increases. At the country level, differences in acceptability for those who blame and do not blame the victim of DVAW are greater in countries with higher levels of GEM (fig 11).

figure ch36533.f1
Figure 1 Graphical representation of significant interaction from three level multilevel analysis of the Eurobarometer 51.0, European Union.

What this paper adds

Reducing the acceptability of all forms of domestic violence against women is one of the fundamental goals of a public health approach to prevent domestic violence. Monitoring the epidemiology of the acceptability of domestic violence against women in our societies is crucial to evaluate outcomes of public education efforts and to design future initiatives. However, almost no research attention has been directed to explore public levels of acceptability of DVAW and to examine its correlates in our societies so that primary prevention efforts can be better informed. After analysing the acceptability of DVAW and its correlates in a representative sample of Europeans of all member states of the EU in 1999, we found that there are still widespread attitudes in the EU, such as victim blaming, which condone domestic violence against women contributing to a climate of social acceptability. Clearly, further efforts to reduce the acceptability of domestic violence against women in the EU are still needed.

Random effects show that the effect of perceived frequency on acceptability (random slopes) was not the same in all localities (0.22, SE  = 0 .06), being greater in those localities with higher levels of acceptability (0.14, SE = 0.04) (covariance intercepts slopes). It suggests that the locality context modifies the association between perceived frequency and acceptability, the slope being steeper in localities where acceptability is higher. Also, we controlled for the effect of interviewer modeling age and sex of interviewer as a random effect. Although not shown in table 44,, respondents had a lower probability of accepting DVAW (β = −0.07, SE = 0.03) to older interviewers and this effect varied across localities (0.09, SE = 0.02). Also, the effect of sex of interviewer varied randomly across localities (0.43, SE = 0.09).

Discussion

This study aimed to analyse through three level multilevel analysis the correlates of acceptability of DVAW in the EU, using national representative data of all member states in 1999 (n = 13 457). At the individual level, those who blame the victim of DVAW because of her provocative behaviour present higher levels of acceptability. Results from this survey show there is a high prevalence of victim blaming attitudes in the EU. Blaming the victims may lead to further portray some violence against women as more tolerable.6,27,44 According to social psychological theory, when victims are believed to cause their own troubles or to get what they deserve the chances for the victims of receiving help or moving away from violence are significantly reduced.17,45,46 These results suggest the need for public education efforts to target prevalent victim blaming attitudes.6,47,48

Policy implications

  • Public policies aiming to reduce the prevalence of domestic violence against women in our societies need to target the levels of social acceptability of different forms of domestic violence as an important factor that contributes either to reduce or to condone and perpetuate violence.
  • Public policies aiming to reduce the acceptability of domestic violence against women are clearly needed. These public education efforts need to transmit the idea that all forms of violence are wrong, and must not be accepted under any circumstances. Reaching zero tolerance must be the final aim.
  • Challenging victim blaming attitudes is a main target for public education initiatives aiming to reduce levels of acceptability of domestic violence against women. Blaming the victims implies acceptance of violence under some circumstances, and reduces the victims' chances of receiving help.
  • Policies achieving a significant reduction of the social acceptability of all forms of domestic violence against women would increase the social costs for perpetrators and contribute to the social control of domestic violence against women.
  • Monitoring changes in levels of social acceptability of domestic violence and its correlates can be an important tool to evaluate public education efforts outcomes, and to help design future prevention initiatives.

Perceived frequency of DVAW is negatively associated with its acceptability. As Klein et al argued, social and personal accountability to take action against DVAW comes from believing that the problem is widespread and of sufficient threat to the community fabric that affects one's own life.6 It follows that an important target for public education campaigns directed to reduce social tolerance towards DVAW is to increase awareness of the alarming pervasiveness of DVAW in our society.

We also found that men who know victims present lower levels of acceptability whereas men who know aggressors present higher levels. The latter is somehow worrisome and suggests the existence of certain acceptance or sympathy for the offender. These positive attitudes towards the perpetrator of DVAW among men may lead to perpetuate, approve or even encourage men's violent acts against women. In this study, 32% of men knew a perpetrator of DVAW in their social circle. Within this group of men DVAW is more acceptable or “understandable”, contributing to condone DVAW in some circumstances and reducing the probability of a known incident to be reported, or help to be offered to the victim. These results controlled for age of interviewer, with respondents having a lower probability of accepting DVAW to older interviewers. Clearly, public education efforts that challenge these attitudes of tolerance and transmit the idea of social responsibility concerning issues of domestic violence are necessary. Breaking the climate of social tolerance would increase the costs for perpetrators and contribute to the informal social control of DVAW.

At the country level, the differences in acceptability of DVAW among those who blame and those who do not blame the victim were greater in countries with higher levels of GEM. According to these results, victim blaming attitudes are pervasive in the EU and become particularly conspicuous in differentiating citizens who tend to accept DVAW from those who do not. Also, this effect is particularly salient in countries more advanced in terms of gender equality, which may help to better understand research showing that societies undergoing changes toward higher egalitarianism often exhibit higher rates of DVAW.34,35,36,37 It follows that challenging victim blaming attitudes emerges as a main target for public education initiatives, comparatively independent of the country's level of gender equality. In summary, a public education effort aiming to reach zero tolerance towards DVAW needs to transmit the clear message that DVAW is wrong, no matter what the reason is, and must not be accepted under any circumstances.49 Reduction of DVAW rates are thought to be related to changing the cultural norms that approve violence.18,19,25 However, as our analysis showed, there are still widespread attitudes in the EU such as victim blaming that condone DVAW, contributing to a climate of social tolerance.17

Finally, the study presents several limitations. Response rates considerably varied between countries and no information is provided by the survey about the characterisation of non‐responders to ensure that non‐response bias does not threaten the validity of the findings. However, additional analyses constrained to countries with a response rate of 40% or more did not substantially change the results presented in this paper. Also, analyses were restricted to valid cases (85% of those initially surveyed), which threatens the generalisability of the study. None the less, recoding variables of interest to increase sample size did not change the results, although it made interpretation of coefficients harder. Also, the survey does not provide information about other potential correlates of acceptability of DVAW such as personal experience of violence and crime. Finally, theoretically meaningful social area aggregations (census tracts, neighbourhoods) and contextual description of these areas (rates of domestic assaults reported to the police, poverty, and other structural conditions) were not available from the survey.50 Clearly, this additional information would help to more accurately estimate the effects seen in this study.

Acknowledgements

We thank Fabio Volante for providing access to Eurobarometer data and technical reports.

Abbreviations

DVAW - domestic violence against women

EU - European Union

GEM - gender empowerment measure

Footnotes

Funding: none.

Conflicts of interest: none declared.

References

1. American Medical Association Diagnostic and treatment guidelines on domestic violence. Chicago, IL: American Medical Association, 1994. 5–6.6
2. Bachman R, Saltzman L E. Violence against women: estimates from the redesigned survey. NCJ‐154348 special report. Washington, DC: US Department of Justice, Bureau of Justice Statistics, 1995, http://www.ojp.usdoj.gov/bjs/pub/pdf/femvied.pdf (accessed 22 Aug 2005)
3. Council of Europe Recommendation Rec(2002)5 of the Committee of Ministers to member States on the protection of women against violence adopted on 30 April 2002 and Explanatory Memorandum. Strasbourg, France: Council of Europe, 2002
4. Hagemann‐White C. European research on the prevalence of violence against women. Violence Against Women 2001. 7732–759.759
5. Kury H, Obergfell‐Fuchs J, Woessner G. The extent of family violence in Europe. Violence Against Women 2003. 10749–769.769
6. Klein E, Campbell J, Soler E. et alEnding domestic violence: changing public perceptions/halting the epidemic. Thousand Oaks, CA: Sage, 1997
7. Krantz G. Violence against women: a global public health issue. J Epidemiol Community Health 2002. 56242–243.243 [PMC free article] [PubMed]
8. Saltzman L E, Fanslow J L, McMahon P M. et alIntimate partner violence surveillance: uniform definitions and recommended data elements. Atlanta, GA: Centers for Disease Control and Prevention, 2002. 85–94.94
9. World Health Organisation World report on violence and health. Geneva: World Health Organisation, 2002. 1–21.21
10. World Health Organisation Violence against women: a priority health issue. Women's health and development, family and reproductive health. Geneva: World Health Organisation, 1997, http://www.who.int/gender/violence/prioreng/en/index.html (accessed 22 Aug 2005)
11. Benson M L, Greer L F, DeMaris A. et al Neighborhood disadvantage, individual economic distress and violence against women in intimate relationships. J Quant Criminol 2003. 19207–235.235
12. Browning C R. The span of collective efficacy: extending social disorganization theory to partner violence. J Marriage Fam 2002. 64833–850.850
13. Cunradi C B, Caetano R, Clark C. et al Neighborhood poverty as a predictor of intimate partner violence among white, black, and Hispanic couples in the United States: a multilevel analysis. Ann Epidemiol 2000. 10297–308.308 [PubMed]
14. O'Campo P, Gielen A C, Faden R R. et al Violence by male partners against women during the childbearing year: a contextual analysis. Am J Public Health 1995. 851092–1097.1097 [PMC free article] [PubMed]
15. Koenig M A, Ahmed S, Hossain M B. et al Women's status and domestic violence in rural Bangladesh: individual‐ and community‐level effects. Demography 2003. 40269–288.288 [PubMed]
16. McQuestion M J. Endogenous social effects on intimate partner violence in Colombia. Soc Sci Res 2003. 32335–345.345
17. Gracia E. Unreported cases of domestic violence against women: towards an epidemiology of social silence, tolerance, and inhibition. J Epidemiol Community Health 2004. 58536–537.537 [PMC free article] [PubMed]
18. Biden J R. Violence against women: the congressional response. Am Psychol 1993. 481059–1061.1061 [PubMed]
19. Goodman L A, Koss M P, Fitzgerald L F. et al Male violence against women: current research and future directions. Am Psychol 1993. 481054–1058.1058 [PubMed]
20. Corrin C. Women in a violent world: feminist analyses and resistance across ‘Europe'. Edinburgh, Scotland: Edinburgh University Press, 1996
21. Council of Europe Men and violence against women. Strasbourg: Council of Europe, 1999
22. Dobash R E, Dobash R. Women, violence and social change. London: Routledge, 1992
23. European Commission Equal opportunities for men and women in the European Union. Annual report 2001. Luxembourg: European Commission, 2002
24. Straus M A, Gelles R J. Societal change and change in family violence from 1974 to 1985 as revealed by two national surveys. J Marriage Fam 1986. 48465–479.479
25. Straus M A, Kaufman Kantor G, Moore D W. Change in cultural norms approving marital violence from 1968 to 1994. In: Kaufman Kantor G, Jasinski JL, eds. Out of darkness: contemporary perspectives on family violence. Thousand Oaks, CA: Sage, 1997. 3–16.16
26. Kruttschnitt C, McLaughlin B L, Petrie C V. Advancing the Federal research agenda on violence against women. Washington, DC: The National Academies Press, 2004
27. Loseke D R. “Violence” is “violence”…or is it? The social construction of “wife abuse” and public policy. In: Best J, ed. Images of issues: Typifying contemporary social problems. New York, de Gruyter 1989. 191–206.206
28. Muehlenhard C L, Kimes L A. The social construction of violence: the case of sexual and domestic violence. Pers Soc Psychol Rev 1999. 3234–245.245 [PubMed]
29. Fagan J. Contributions of research to criminal justice policy on wife assault. In: Besharov D, ed. Family violence: research and public policy issues. Washington, DC: American Enterprise Institute Press, 1990. 53–81.81
30. Conger R D, Elder G H, Lorenz F O. et al Linking economic hardship to marital quality and instability. J Marriage Fam 1990. 52643–656.656
31. MacMillan R, Gartner R. When she brings home the bacon: labor‐force participation and the risk of spousal violence against women. J Marriage Fam 1999. 61947–958.958
32. Smith M D. Sociodemographic risk factors in wife abuse: results from a survey of Toronto women. Can J Sociol 1990. 1539–58.58
33. Straus M A, Gelles R A, Steinmetz S K. Behind closed doors: violence in the American family. Garden City, NY: Doubleday, 1980
34. Pallitto C C, O'Campo P. Community level effects of gender inequality on intimate partner violence and unintended pregnancy in Colombia: testing the feminist perspective. Soc Sci Med 2005. 602205–2216.2216 [PubMed]
35. Schuler S R, Hashemi S M, Riley A P. et al Credit programs, patriarchy and men's violence against women in rural Bangladesh. Soc Sci Med 1996. 431729–1742.1742 [PubMed]
36. Yllo K. Sexual equality and violence against wives in American states. J Comp Fam Stud 1983. 14676–686.686
37. Yllo K, Straus M A. Patriarchy and violence against wives: the impact of structural and normative factors. Journal of International and Comparative Social Welfare 1984. 11–13.13
38. European Commission Europeans and their views on domestic violence against women. Eurobarometer 51.0. Brussels: European Commission, Directorate General X, 1999. http://europa.eu.int/comm/public_opinion/archives/ebs/ebs_127_en.pdf
39. United Nations Development Programme Human development report 1999. New York: Oxford University Press, 1999
40. Merlo J, Chaix B, Yang M. et al A brief conceptual tutorial of multilevel analysis in social epidemiology: linking the statistical concept of clustering to the idea of contextual phenomenon. J Epidemiol Community Health 2005. 59443–449.449 [PMC free article] [PubMed]
41. Rasbach J, Steel F, Browne W. et alA user's guide to MLwiN. Version 2.0. London: Institute of Education, University of London, 2004
42. Raudenbusch S W, Bryk A S, Cheong Y F. et alHLM 6 Hierarchical linear and nonlinear models. Lincolnwood, IL: Scientific Software Interantional 2004
43. Merlo J, Yang M, Chaix B. et al A brief conceptual tutorial on multilevel analysis in social epidemiology: investigating contextual phenomena in different groups of people. J Epidemiol Community Health 2005. 59729–736.736 [PMC free article] [PubMed]
44. Koss M P, Goodman L A, Browne A. et alMale violence against women at home, at work, and in the community. Washington, DC: American Psychological Association, 1994
45. Lerner M J. The desire for justice and reactions to victims. In: Macaulay J, Berkowitz L, eds. Altruism and helping behavior. New York: Academic Press, 1970. 205–229.229
46. Weiner B. A cognitive (attribution)‐emotion‐action model of motivated behavior: an analysis of judgments of help giving. J Pers Soc Psychol 1980. 39186–200.200
47. Banyard V L, Plante E G, Moynihan M M. Bystander education: bringing a broader community perspective to sexual violence prevention. J Community Psychol 2004. 3261–79.79
48. Sabol W J, Coulton C J, Korbin J E. Building community capacity for violence prevention. J Interpers Violence 2004. 19322–340.340 [PubMed]
49. Salazar L F, Baker C K, Price A W. et al Moving beyond the individual: examining the effects of domestic violence policies on social norms. Am J Community Psychol 2003. 32253–264.264 [PubMed]
50. O'Campo P. Advancing theory and methods for multilevel models of residential neighborhoods and health. Am J Epidemiol 2003. 1579–13.13 [PubMed]

Articles from Journal of Epidemiology and Community Health are provided here courtesy of BMJ Group
PubReader format: click here to try

Formats:

Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...

Links

  • MedGen
    MedGen
    Related information in MedGen
  • PubMed
    PubMed
    PubMed citations for these articles

Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...