Prevalence and Risks of Physical and Sexual Violence against Women by Nonintimates: An Exploratory Study across Nine Countries

Abstract

Research on violence against women involving nonintimates is a relatively understudied area. This research note presents data for nine countries participating in the International Violence Against Women Survey (IVAWS), including estimates of physical and sexual violence perpetrated by acquaintances and strangers and multivariate models assessing the effects of respondent age, source of income, current relationship status, and previous victimization experiences. Results show that there is considerable variation among countries in the prevalence and predictors of nonintimate violence. This suggests that women in different countries may be exposed to nonintimate violence under different circumstances. These results underscore the need for in-depth research to better understand the broader cultural, social, and economic contexts in which women are exposed to a range of violent experiences.

Keywords:

• violence against women
• nonintimate violence
• cross-cultural
• risk factors

Acknowledgments

We would like to acknowledge the financial support of the agencies in the eleven countries participating in the International Violence Against Women Survey (see Johnson, Ollus, & Nevala, 2008, pp. 195–197). The work of the first author was supported by the Swiss National Science Foundation in the form of a fellowship for prospective researchers, project PBLAP1-131842. The authors would like to thank Professor Heidi Scherer for helping us with exploratory hierarchical generalized linear modeling.

Notes

1. Country-level characteristics (e.g., gross domestic product, fertility rate, expected years of schooling, female labor force participation) were considered for inclusion in multivariate models; however, preliminary statistical analyses predicting victimization using hierarchical generalized linear modeling were not stable with nine countries.

2. Greece and Italy participated in the IVAWS but were excluded from the present analyses, Greece because of its small sample size (n = 479) and Italy because the questionnaire was altered to a degree that comparison with other countries is compromised.

3. The project coordinator in each country decided which language and mode of questionnaire administration were employed based on practical and logistical factors, available technology, and budget.

4. Prevalence within the last year would have been preferable to more accurately assess risk factors but the number of cases in some countries was too small to allow for reliable analyses.

5. Across countries, almost 19% of victims reported sexual abuse by a father or a brother, with less than 1% reporting abuse by a mother or a sister. The majority of women indicated they were abused by another relative (24%) or someone they knew (62%), but very few provided further details regarding their relationship to these perpetrators. It is thus possible that some of these perpetrators were female.

6. Although not presented in Table 4, household quartile was included in multivariate regressions. Compared to the reference category (i.e., higher income quartile), the lowest quartile was a marginally significant risk factor in Australia and Switzerland (p ≤ .10) for acquaintance violence; it was marginally significant (p ≤ .10) in Denmark and Hong Kong for stranger violence.