Browsing through the blogosphere can offer some interesting insights into all sorts of situations. One of my pet fascinations (probably to the extent of it being extremely unhealthy) is causation. So much legislation is passed which is based on an assumption of causation that may or may not be right. Often this assumption is based on statistics and that can be very misleading. There are many reasons for this.
First and foremost is the creation of the statistic which will usually involve the selection of the group to be analysed (people in the case of a poll but it can be almost anything such as soil samples). Unless the group choice is both neutral and comprehensive, the statistics will be inaccurate.
Then comes the interpretation of the statistics. Clearly I am in the area of crime in this blog so the obvious example is to look at the statistics that indicate that crime rates are dropping. These can be interpreted in two ways: crime rates really are dropping – thanks to the perception that the police are useless less crime is reported. In my view the truth is probably a bit of both but that opinion is pretty worthless (how could I really know?).
Lastly there is the determination of the cause based on these statistics. If crime really is dropping, the next stage should be to try to find out why. However, in the frantic world of politics where people are trying to combine two activities that are mutually exclusive (doing what is right for the nation plus remaining electable) the reaction is simple: if it ain’t broke don’t mend it. Actually not a bad reaction since “mending” always carries “unexpected consequences” but that is another subject.
If you think, however, that the cause is poor policing you will up the police anti by employing more and giving them greater powers. This could well prove a disaster if you have misread the causation: it could create an even greater divide between the people and the police but do nothing toward reducing crime.
So it was with the greatest pleasure that I stumbled on a post by Reality Swipe canned “The Rise and Fall of a Violent Society“. In short it demonstrates (and I feel conclusively) that when youngsters ingest – from the atmosphere or from painted objects, etc. – lead there is a fundamental change in the way in which the brain develops that means they will be more likely to become violent in later years. Read it for yourself if you want to be convinced and be grateful that we now live in a world where lead is no longer used so widely (especially in motor fuels – which resulted in high lead levels in the atmosphere – and in paints). Thus we do not need to take further action in this case but it clearly shows the need to find causation before acting to correct a problem.
Obviously removing one cause of violence will not remove all violence so now we need to look at what else needs attention. Since I was complaining about statistics and now I am accepting that statistics enabled us to uncover this causation I should add that these statistics come from the right group and are comprehensive. The one crime that is bound to be reported to the police is murder (unless carried out so discretely that murder is not suspected in which case it is hardly an example of a violent crime) so we may accept the reductions in the murder rate as an accurate figure. Likewise the drop in the level of lead in the environment is not disputed although more needs to be done to lower it further.
Let me finish with a quote from Reality Swipe’s blog. “Like most good science, this has fallen on deaf ears. There is minimal political push around the world for this to be addressed. Most politicians, like good old Rudy Giuliani, are happy claiming that their policies are the reason for this drop in crime . . .”