24 February 2010

Death by numbers...

Must scoot out early into the rosy-fingered dawn, so I thought I’d soften you all up with some graphs constructed from statistics released yesterday in the Scottish Government’s Homicide in Scotland 08/09 numbers. These sort of statistics are important – as they capture quantitatively that which escapes ordinary experience and gives the lie to impressionistic accounting about the social prevalence of murder and the modal characteristics of imaginary victims. These concrete numbers give us something far more substantial to go on. Do click on the graphs themselves for a more focussed view. Later on, I'll be returning to the publication and examining it in a little more detail.

Firstly, I present Figure 1, which tracks the total number of recorded victims of homicide in Scotland since 1999-00.

Secondly, Figure 2., which illustrates the overall division of killings by gender of the victim in the last half-decade, 2004/05 - 2008/09.

Thirdly, for Figure 3. I've taken the correlated figures about the age of the victim and presented that alongside the total homicides recorded.

Finally, in Figure 4, I focus on the latest year of record, 2008/09. Alongside total figures for that year, divided along gender lines, I've also presented how the numbers breakdown in respect of gender and age of victim.


  1. How do the skills of the paramedics to 'scoop and run' to A/Es staffed by competent medical and nursing staff, increasingly skilled at emergency life support, and the further skills of accomplished surgeons and aftercare affect the numbers?

    As with the shocking events in Afghanistan etc. emergency transportation and medical skills probably alter the potential death rates dramatically?

    I know of one large district hospital where the survival rates following heart attack were excellent compared to a large inner city where the survival figures were dreadful. If, however, you are within walking distance of a hospital the likelihood of all heart attack patients arriving to either survive or die are very high - therefore the apparent survival rate is poor. If, however, you live miles from the nearest hospital it is likely that only those who can survive will arrive - thus the apparent survival rate is high! Statistics - eh?

    Perhaps we need a graph for the near misses?

  2. That is an interesting question Clarinda - and the obvious answer is that I haven't a clue and I imagine that the office-bound Government statisticians ain't got one either. Particularly given the instance of drunkenness and the use of sharp instruments - the Idiot Murder paradigm - the modal sort of killing suggested would, as you say, seem to leave a possibility for the intervention of paramedics.

    You also raise an important point about the limits of statistics and quantitative research. It leaves important variables uncaptured - it doesn't account for the possibility you mention - and thus any causal statements about a complex range of issues are bound to be simplifying, a bit gauche. That said, the Government Statisticians aren't really falling into that sin here - they're just totting up and presenting comparable figures, year on year.