A/HRC/7/6 page 11 and disjunctions between research and policy definitions of intimate partner violence (see below). If we had multiple international studies, covering girl and womanhood, and the contexts of employment and education, we may conclude that sexual harassment is the most common form of violence against women. A recent German prevalence study, for example, found rates for sexual harassment four times greater than for physical or sexual violence from a current partner.14 38. The prevalence survey has emerged as a route to establish more accurate estimates of the scale of violence against women, the extent of which is generally assessed through two different measurements: lifetime prevalence and that for the past 12 months. Presented as a rate, prevalence establishes the proportion of the female population that has experienced violence. The frequently cited headline figures, such as “one in four women have suffered intimate partner violence”, are lifetime prevalence rates. If we are to see falling rates of violence as a result of new policies and interventions, it will be detected through decreasing 12-month prevalence rates. Such rates are more reliable where research samples are as representative of the entire population as possible. 39. Some forms of violence against women are single events; other forms may be recurrent, with their gravity and impact a combination of intrinsic harms and repetition. The amount of violence - variously understood as incidents/frequency/chronicity - is a required layer of measurement for forms of violence which tend to be “courses of conduct”: intimate partner violence, stalking and sexual harassment are the most obvious examples here, but sexual violence/abuse is also often repeated. There has been much less methodological development in measuring sexual violence compared to intimate partner violence and female genital mutilation. Ongoing work will be needed to ensure comparative data since there is no consensus in research or national law about either the boundaries between rape and other forms of sexual violence or what constitutes rape. The latter is variously defined in terms of force, lack of consent, and coercive circumstances, each of which has different implications for the construction of questions and data analysis. 40. Measuring trafficking for sexual exploitation has specific challenges: not only are the practices more hidden, but the international definition is complex, as difficult for researchers to operationalize as for State officials to prove. Methodological innovation is beginning, however, with some questions included in surveys on violence against women15 and studies of irregular migrants. 41. An important development in research on female genital mutilation was the transnational agreement reached on a definition that in turn permitted more accurate documentation and new 14 The study also found higher rates of intimate partner violence for migrant women and higher rates of all forms of violence against women for prostitutes, prisoners and refugees/asylum-seekers (Hageman-White, 2005: 305). 15 African Child Policy Forum, 2006, Violence Against Girls in Africa: A Retrospective Survey in Ethiopia, Kenya and Uganda.