There were many differences observed between the two periods in respect of the kinds of violations reported. However, there is one methodological comment to be made here, and this relates to the manner in which violations were classified in the two periods. In 2002, there was a much greater tendency to classify physical abuse as torture, whereas in 2008 many cases of physical violence that may well have been torture were classified as Assault with Intent to do Grievous Bodily Harm or Aggravated Assault. This is a common problem in human rights data bases, and a future task must be to re-code all violations into a common format such as HURIDOCS. However, despite this, and accepting that violations occurred irrespective of the classifications, it is possible to see the similarities and differences between the two election periods. Figure 6 Comparison of alleged violations: 2002 versus 2008 As can be seen from Figure 6, there are very different patterns observed between the two periods. With the exception of torture, for reasons indicated above, virtually all violations were more significantly more frequent in 2008. There was a 25% drop in torture in 2008, but a 55% increase in the number of assaults, but, as seen earlier, Assault [GBH] was reported more frequently than Assault in 2008, which suggests that torture may well have been more common in 2008 than the data indicates.43 Also it is evident that there were significant increases in abductions, intimidation, 43 73% of the cases in 2008 reported Assault [GBH], which suggests that torture went up rather than down in 2008 in line with the general trend for there to be more frequent violations in 2008 than 2002. 2008 versus 2002: Assault [p=0.001]. 2002 versus 2008: Torture [p=0.001].