There are statistics on victims of crime, reported crime and punished crime but no statistics on crime itself. This is _ obviously _ because a lot of crime goes unreported. As a result, people use statistics on reported crime and punished crime (convictions) as a substitute. It may seem reasonable, but it is not.
We know, for example, despite a surge in reports of sexual assault, that the underlying rate has not fluctuated much. General surveys of the population about whether people had been a victim of a crime indicate that. It is just that changes of attitude in police forces, the courts and society in general make reporting a less traumatic experience, so more victims are willing to report. Contrarily, as insurance companies lifted their excess, reportage of theft and malicious damage fell, but the underlying rate probably remained the same. In short, reporting figures are rubbery. Now it seems that conviction figures can also be as misrepresentative.
Figures last week show that the number of people charged with drink-driving after a random test in the ACT is 10 times that of NSW. Shock, horror. Canberra a city of reckless drunks. Time for some vote trawling and pleas for extra police resourcing. In fact, however, this shows more about police behaviour than drivers’ behaviour.
In 1993-94 ACT police did 96,626 random breath tests. As a result they prosecuted 1083 people. In rough terms one person per hundred were prosecuted. In the same year in Newcastle, a city of comparable size, NSW police did about the same number of tests _ 102,868. However, their testing resulted in only 106 people prosecuted. That is, about one person in a thousand, or 10 times fewer than the ACT rate.
It is scarcely, conceivable that the ACT drink-driving rate is 10 times that of Newcastle. During my visits to Newcastle I saw comparable hoonism in cars and public drinking as in Canberra. There was nothing to suggest its drivers or drinkers were radically different from Canberra’s.
Do Canberra police do a lot more tests than in NSW? Answer no. They do slightly more at 96,000 tests in a population of 280,000 compared to an overall rate of 1.7 million tests in 6 million population.
The difference comes in police tactics. NSW police go for general deterrence. They randomly test in highly visible places with highly visible booze buses. They hit every driver, not just a selection. It does not matter that they set up at times and places where there are likely to be very few drink drivers on the road. They even test in the pouring rain.
They then reinforce the message with television advertisements which dwell on the highly visible road-block style. The message is: anywhere; anytime; there is no escape.
ACT police tend to hit certain hot spots at hot times _ instances when drink driving is likely to be more prevalent. Typically, in times when there are few drivers on the road. They get a lot of convictions. Unlike most of NSW _ in particularly the larger cities _ the convictions get reported in the press. The Canberra Times publishes every one. In Canberra shaming is a big part of deterrence.
There is no telling which method is more effective. A control-group statistic is not available for the ACT, though it is for NSW. That statistic is the number of prosecutions for drink driving obtained through methods other than random testing _ accidents and drivers picked up committing other offences.
In Newcastle, a further 27,000 tests resulted in a further 436 convictions. That figure is not available for Canberra _ a pity, because it would be a truer indication of the rate of drunkenness behind the wheel than the random testing result, which as we have seen is not so “”random” after all. NSW police admit that when they move random testing to high-drink zone targets, they net six times the number of prosecutions if.
The point is that the over-reaction to ACT figures is unjustified. It is probably an attempt to get more money for police or serves some political law-and-order agenda. If there is to be any more money spent it should go on collecting better statistics to make an informed decision about deterrence.