Tool 5: Places of Deprivation of Liberty and Gender Box 7: Data collection for contextual understanding in the USA After an influential report by Human Rights Watch in 2001 highlighted the endemic problem of prison rape and sexual slavery in US prisons, the Prison Rape Elimination Act was passed in 2003. A specific recommendation of the Human Rights Watch report was that the National Institute of Corrections “should make an effort to collect, maintain and disseminate data relating to prisoner-on-prisoner sexual abuse”, in recognition of the fact that no nationallevel data were collected (Human Rights Watch, 2001, Section III: Recommendations to the National Institute of Corrections). The Prison Rape Elimination Act mandated the national Bureau of Justice Statistics to “carry out, for each calendar year, a comprehensive statistical review and analysis of the incidence and effects of prison rape”, and thus the Survey of Sexual Violence was created (Rantala, 2018: 4). Regular monitoring of the data facilitated a better understanding of sexual violence in prisons and led to further recommendations being made. In 2013 the Survey of Sexual Violence was renamed the Survey of Sexual Victimization, to move further in line with national standards developed as a result of the Prison Rape Elimination Act. Definitions were updated and questions about inmate-on-inmate sexual harassment were added. Demographic information was widened to include transgender and intersex categories. Data gathered from the surveys continue to feed into policy-making. The Bureau of Justice Statistics uses uniform definitions for each sexual act and investigative outcome. Each sexual act is classified by the type of perpetrator (i.e. inmate or staff) and type of act. Some examples of data collected on sexual victimization in places of detention include:  percentage of prisoners reporting one or more incident of sexual victimization by another prisoner since admission  percentage of prisoners reporting one or more incident of sexual victimization by a staff member since admission  data disaggregated and analysed by gender (female/male/transgender) and ethnicity  data on those in both short-term detention facilities (jails) and long-term centrally managed detention facilities (prisons). Sources: Human Rights Watch (2001) “No escape: Male rape in U.S. prisons – Summary and recommendations”, www.hrw.org/reports/2001/prison/report1.html#_1_6 (accessed 17 October 2019); R. R. Rantala (2018) “Sexual victimization reported by adult correctional authorities, 2012–2015”, July, Washington, DC: Bureau of Justice Statistics of the US Department of Justice. 4.2. Guidance for places of deprivation of liberty Staff selection and ongoing training While in some countries places of deprivation of liberty for women do have men working directly with the women and managing the institution, the Mandela Rules (Rule 81) are clear that a female institution should be led by a woman and there are certain areas of the establishment which men should not enter. This has implications for recruitment of staff. Some countries have introduced targets and quotas for the number of female staff required. Training for staff has historically been the same whether the staff member is due to be deployed to a place of deprivation of liberty detaining women or men, and staff could be moved between such institutions without any further training. Some states have introduced specific training for working with women in detention as being mandatory before deployment to work with women (see Box 8). This approach is important, because it asks the staff member to think from the outset (or fundamentally rethink, if they are being redeployed to a women’s facility from a men’s facility) about what it means to work with 31