11 reports and the reports of media monitoring organizations. However, the MDC was not blameless here, but still was 4 times less likely to indulge in hate speech. Table 14. Hate speech. Organization % of constituencies reporting. ZanuPF 57% MDC 14% Other 4% Overall, the data derived from the reports of the community monitors does not support the remarks of the ZRP in any but the most trivial fashion. Overt physical violence may be reduced, but almost every other indicator is high. The direction of the violence is overwhelmingly due to ZanuPF rather than the MDC, although the MDC are not wholly blameless. The data received by the MDC corresponds to much that is being reported in the press and it is clear that the election climate is far from satisfactory in respect of political violence. Militia The most frequently reported association with the presence of militia in a constituency or district has been with physical violence. This has been highly contested by the government, but, nonetheless, a number of influential reports have implicated militia, and, particularly the youth militia, in the perpetration of political violence. Several reports have also drawn attention to the relationship between political violence and elections, and these reports have also implicated the role of the youth militia in the perpetration of gross human rights violations. The current data has allowed us to examine a number of aspects of the election climate with respect to militia, although it does not allow us to specify whether the reports relate to militia as war veterans, youth militia or party youth wings. The most important relationship to examine is the relationship between militia and violence, since this has been the most frequently relationship in the press and in human rights reports. It has drawn comment from the African Commission for Human and Peoples’ Rights in the report of the Mission in 2002, and was commented upon in the reports of several observer groups to the Presidential Election. As is seen in Figure 2 [see over] there are marked differences between constituencies that report the presence of militia and those that do not. Figure 2 contrasts two measures: the Electoral Irregularity Rating and a Violence Score. The Electoral Irregularity Rating needs little comment, but some brief words are necessary about the Violence Score. This latter score was derived from 5 indices of violence in the reports of the community monitors: physical violence, gender-based violence, intimidation, inter-party violence, intra-party violence, and forcing people to chant political slogans. As is seen from Figure 2, both Electoral Irregularity Ratings and Violence Scores are higher in constituencies that report the presence of militia. This suggests that there are two ways to examine the influence of the militia on the electoral process National Constitutional Assembly: Election Climate Reports.