Crime statistics actually

'Understanding what crime statistics actually mean is crucial' (Abercrombie and Warde). Elucidate. As Coleman and Moynihan argue 'the data we have discussed should never be accepted as given, but are products, socially constructed, often reflecting the dimension of power' (1996:142. ) It is crucial to understand exactly what crime statistics mean as different data varies in its reliability and validity. Validity of the statistics refers to the issue of whether the agents measure exactly what they say they are supposed to measure.

Reliability equates to whether or not this measuring is done in a fair, unbiased, accurate and consistent way. It is necessary to value crime statistics like this, as there are three main types of data, which have varying degrees of reliability and validity. These are firstly Official Statistics most prominently the police Criminal Statistics England and Wales used not only by people within the criminal justice system and those studying crimes but also by politicians, the media and the general public.

This leads to inaccuracies in reliability through police performance and reaction, embarrassment and fears of consequences of reporting. They also have a tendency to over-represent working class crimes and under-represent corporate and government crimes including those committed by themselves, the police. Secondly, there is Victimisation Surveys- namely the British Crime Survey (BCS); these identify the victim of the crime as opposed to the offenders. It has been contested that they have a greater validity and reliability as they show mush criminal behaviour that goes unreported and therefore unrecorded.

Lastly, there are Self-Report Studies, which ask directly about an individuals involvement in rule breaking and crime. With these studies, in particular, it is important to work out what the objectives of them are in order to find out their validity, reliability and meaning. Nevertheless, there is a shadow that is cast over all types of crime statistics which is crucial to their meaning and this is the 'dark figure'- unrecorded crime and undetected offenders.

Hence, the reason these quantitative statistics were put together, by whom and the way in which there are measured is crucial in establishing their meaning and therefore their validity and reliability. It is important to question the meaning of Official Statistics as their validity and reliability are very questionable as they are concerned with who committed the crime rather than the total number of offences and they only represent a segment of crime known to the police.

However, even if a crime is reported the rules say there can only be one offence per victim and where there is several offences only the most serious is counted plus where there is a continuous series of offences only one offence is counted. Therefore, as many crimes go undetected and even if they are many are unreported resulting in a gross under-estimation. This can be shown by the fact that the Official Statistics are supposed to be realistic in the area of homicide yet, this is questionable due to the number of alleged murder victims of long ago and the vast number of 'missing persons'.

Likewise, it can be shown that these statistics only touch the tip of the 'iceberg' shown by the above material, this 'dark figure' can be further compounded by the work of Sutherland, a member of the Chicago School. He asserted that white collar crime (pollution, health and safety at work plus unethical business practise e. g. extortion, duress) was not only widespread but a large measure of them went unreported, due to their political, economic and social meaning.

Furthermore, the British Crime Survey (below) 1988 estimated 60,000 sexual offence incidents occurred against women in1987, 13,000 of which were reported by the police. This shows that actual sexual offences where 4. 6 times higher than those reported; therefore they can't possibly be argued as realistic. In conclusion the meanings of the Official Statistics is very important as they greatly under-estimate offences and have a very high 'dark figure', therefore the status they are given is not warranted, as they are invalid and unreliable.

Victimisation Surveys especially the British Crime Survey 'present a clearer and more complete picture of some forms of criminality… but they do not give the whole picture. ' (Sparks 1977:227. ) Although levels of reliability and validity vary according to the type of crime, from a social aspect these surveys are very reliable as they give a more accurate estimation of the true crime rate and any changes in it as well as the police's recording practises; therefore holding them accountable. They also provide an important social function of acknowledging, the usually ignored victim.

The BCS as the most recognised Victimisation Survey has been useful in showing the inadequacies of the Official Statistics as shown above and give good insight into public attitudes to crime and the criminal justice system. However, the Victimisation Surveys are far from perfect as their major flaw is that they can only measure criminal incidents where there is an identifiable victim, this therefore excludes victimless crimes such as drug offences. The incident must also be perceived and definable to be recorded.

Therefore, Victimisation Surveys are best suited to measure ordinary crimes like theft, although they are better at recording rape etc. the Crime Statistics are as accurate as Local Surveys which can be designed specifically for this purpose, likewise, they do help to quantify the 'dark figure' but not to a reliable extent. This can be shown by the first BCS on the issue of sexual offences, as these were clearly under-estimated as victims were reluctant to tell the interviewer and it was also possible that the assailant was in the room.

In conclusion, the Victimisation Surveys show that it is critical to understand what statistics actually mean as they research 'the social meaning of victimisation' (Sparks, 1977:231) therefore from a victims perspective they have they own unreliability and invalidity. The meaning of Self Report Studies is crucial in understanding the data as they entail asking people directly of their crime involvement this can lead to low validity and reliability, as people may not confess the truth.

However, these reports are very important as they give the perspective of the offender (especially delinquent behaviour) and show that rule breaking behaviour is more widespread then previously thought. Nevertheless, this is also a limitation of the studies as the behaviour of delinquents produces a narrow, bias outlook. Furthermore, these studies have many methodological issues; not least their avoidance of the controversy over the definition of what crime is, it had been said by Box, S that the adoption of a legalistic definition of crime is a political decision.

Likewise, the issues of validity and reliability can't be avoided, as the construction of these studies has to be scrutinised too trivial with overlapping question confusing the meaning. Hence, these studies are still haunted by the 'dark figure' which is ironic as they were set up to counter act this in the Official Statistics; corporate and governmental violations are not exposed as well as hidden crime, of sexual, domestic and physical violence. These studies show the importance of understanding their meaning in terms of objectives as different implications arise.

This can be shown by the results being accumulated in a different context as they have the objective of exploring epidemiology and testing theories of delinquency and deviance. This means that they provide very sociological research data but have less meaning in terms of official, political criminal statistics. It is important to analysis all crime data to ensure that it's true meaning is understood; this is because as shown above no research, survey. Study or data can contain all knowledge and be totally reliable and valid.

Therefore, in crime statistical analysis, all three types of research laid out above need to be analysed to find their exact meaning through the context they were written in (who asked for them and why) and the objective they were trying to achieve. Each of the three types of data have their own bias and flaws as none of them completely unearth the 'dark figure. ' However, it is possible that these three types of data long with Local Surveys (to unearth the 'dark figure') put together can produce the fairest, most accurate, unbiased, realistic, valid and reliable results.

This is because the Official Statistics focus on the police, the Victimisation Surveys on the victim and the Self-Report Studies and the offender, triangulated they can produce and aggregate to a medium result, neither extremist or institutional, although their views need to be considered. The context the data is written in is very important as they sometimes have a broader political, economic and social aim and can tell you more about the people who wrote the study than from the statistics and their objectives.

Likewise, the politicisation of crime and criminal justice and the demands for accountability must also be considered as this can lead to bias and blurred objectives when in comes to the research and its conclusions. Furthermore, official crime data shows that many of the claims that politicians make about law and order should not be taken at face value as much of their information is either bias and unreliable or completely fabricated.

Therefore, when researching crime statistics it is crucial to look at who commissioned them, the objective they are aimed at, what political, economic and social context they were written in and whose perspective were they from as all these questions lead to the real meaning of the crime statistics which crucial to identifying their validity and reliability. 


Abercrombie N, Warde A et al, 2000, Contemporary British Society, Polity Press Coleman C and Moynihan J, 1996, Understanding crime data, Open University Press Sparks R, Genn H, Dodd D, 1977, Surveying victims, Wiley Walklate S, Gender,2001, Crime and Criminal Justice, Willan