banner supportlinkscontactaboutarticlesmain

Crime Statistics: Methodology

Currently in England and Wales there are two ways of measuring crime and, consequently, leading to different results. Not only does this make our life hard, more importantly it allows politicians to say very different things. We have already shown there is a considerable difference between the two methods - one says violent crime has doubled, the other says it has halved. So read on, for a comparison of the two.

Police Recorded Crime
Police recorded crime is a list of categories outlined in the Home Office Counting Rules, known as ‘notifiable offences.’ The main categories are: violence against the person, sexual offences, robbery, burglary, theft and handling stolen goods, fraud and forgery, criminal damage, drug offences and an ‘other offences’ catch-all that includes offences as diverse as riot and assisting suicide.

In April 1998 many new offences were added including many less violent crimes, including those that are dealt with by summary offences. No distinction is made between the new offences and old, making comparing the statistics before and after this introduction rather difficult.

The National Crime Recording Standard (NCRS) was introduced in April 2002 and requires that the police record a crime if ‘the circumstances as reported amount to a crime as defined by law and there is no credible evidence to the contrary’. The record will remain unless evidence emerges to disprove that a crime has occurred. The introduction of the NCRS in 2002 had the effect of increasing recorded crime by 10% almost overnight simply through a change in recording practices.

Any statistics based on police-recorded crime will obviously exclude crimes that are not reported to the police. And some high volume crimes such as shoplifting are intrinsically difficult to measure in a consistent way – the discovery of missing goods may be attributed to one criminal offence or many. Also, crime is a product of the extant laws and where those laws, their enforcement, or the public’s tendency to report crime changes over time, then the volume of crime will change as a consequence. Furthermore, the victims of some crimes may not trust the response of criminal justice agencies and fear the added trauma of pursuing the case in court. Thus, certain serious crimes remain very under-reported - for example, domestic violence, rape or organised crime.

Various ideas have been put forward to meet it. One is to exclude more minor offences – for example offences recorded as violence but where no-one is injured – so as to focus on the sort of crime that people are most concerned about and which may be less susceptible to fluctuations in recording practice. This is something we are going to explore over the next few articles on this subject.

The British Crime Survey (BCS)

The BCS came about as a result of dissatisfaction with the comprehensiveness of statistics compiled from police records. It was first conducted in 1981 and is produced annually since 2001 by the Home Office. The BCS is an annual survey of about 40,000 people.

It asks individuals whether they have been a victim of certain crimes over the course of the previous 12 months. The main offences covered by the BCS are: vandalism, burglary, vehicle-related thefts (including bicycles), other household thefts, theft from the person, assault, wounding and robbery.

The BCS focuses on crime against an individual, thus eliminating all crime against a business or organisation, including fraud. It fails to take into account "victimless" crimes such as drug offences and crimes such as murder where the victim cannot, for obvious reasons, be interviewed. Rape and other sexual offences are not included, an acknowledgement that many respondents would be unwilling to disclose this information. Crimes against people under the age of 16 are also excluded - removing large numbers of crimes that are common among this age group, such as mobile phone theft and child abuse.

Furthermore, with a survey size of 40,000, some of the more serious crimes will be under-represented. For example, in England and Wales, there were 15,500 serious injuries that endangered life in 2005 reported to police (this particular crime has quite a high liklihood of being reported to the police). With a population of 54,000,000, one in every 3500 of us were victims. This means in the BCS sample size, one would expect to have 11 or 12 positive responses for this crime. Just with statistical fluctuations, let alone a poorely designed survey, one could easily have 6 reports of the crime in the survey. The point is, that with more serious crimes - those that we are far less likely to be victims of, than say mobile phone theft, the BCS is fundamentally flawed.

Which one is best?
BCS crime figures have been going down since 1995, after an upward trend since 1981.
Police recorded figures went down from 1991 until 1997, but have been going up since then.
In other words, the BCS indicates that crime rose under the Conservatives and dropped steeply under Labour, while recorded crime fell under the Conservatives and rose under a Labour government.

Police recorded crime is very susceptible to changes in recording practice which can greatly effect the data collected, the BCS has remained largely the same for 20 years. This, however, is a function of government policy more than it is anything else. The BCS is probably better at indicating the level of crime for less severe offences, such as most property thefts and minor assualts, but most serious offences are best reported by the police. There are of course some crimes that are under-reported in both sets of statistics.

In August 2004, Michael Howard stated “crime has risen by almost 850,000 in the last five years”. In July 2004, Tony Blair wrote “crime has fallen by 25%”. Who would you believe?

Related links:
Statistical Commission report on statistical methodology

Post a comment