Evaluate the accuracy of official statistics of crime.
Statistics provided by the police, courts, and other law enforcement agencies tend to surround the basis for many theories of crime. These statistics tend to consistently portray some groups as more involved in crime than others. Such groups include members of the working class, young people, males, and members of some ethnic minorities. Some sociologists, like Merton, Cohen and Cloward and Ohlin have taken this idea at face value, proceeding to give different explanations as to why this is the case. However, others question the reliability of these figures upon which some theories rely on so heavily.
The official British crime statistics are published annually, providing anyone concerned with information on the total number of crimes ‘known to the police,’ and the social characteristics of those convicted of offences, for example age and ethnicity. The first type of information allows for comparison between crimes over the years, with the latter providing the basis for a number of criminological theories. The former type of statistics demonstrate a long term trend of rates remaining quite low until the 1950s, and increasing rapidly for the most part from then. It is worth noting that from 1998-9 the method of recording indictable offences changed, leading to a considerable increase in recorded indictable crime according to Maguire (2002). Other changes have also had an effect on these figures, and Mike Maguire (2002) does point out that crimes such as tax evasion and benefits fraud are not recorded as they are dealt with administratively — not through prosecutions. Additionally, crimes recorded by the police but not dealt with by the Home Office aren’t included in the main statistics, which account for an estimated extra 80,000 offences each year.
There is also evidence to suggest a “dark figure” of unrecorded crimes, as obviously not all will be recorded. In order for this to happen, it must firstly come to someone’s attention that a crime has occurred. Secondly, they must report it to the relevant agency, which must be willing to accept that the law has been broken. This presents a degree of difficulty when trying to accurately record crime, as not every victim knows they have been wronged. For example where tax evasion occurs, the victim is the community as a whole, rendering this type of crime difficult to uncover without investigation. It is much easier to estimate criminal activity that victims are aware of, but haven’t reported to the police through victimisation studies.
Victimisation studies are conducted with an aim of overcoming the limitations of police statistics, and are now regarded as at least as useful as the latter. The 2001/2002 British Crime Survey (BCS) found that a meagre 42% of crimes they were informed of were reported to the police. This figure was considerably variable depending on the type of crime; from 6% of vehicle thefts to 74.3% of common assaults going unreported. This disparity between different crimes demonstrates the lack of reliability found within police figures. The BCS also looks at the reasons why crimes go unreported, with 72% of those who didn’t report attributing it to the belief that the offence was too trivial, or that the police couldn’t do anything. Around 3% mentioned a fear of reprisals, and 2% mentioned fear of dislike of the police as reasons not to report. These figures back up the ideas of Left Realists, who strongly advocate improved police relations with the public.
The disparity in reporting rates can be particularly useful when considering what can be perceived as a significant rise in crime. For example, in 1981 and 2001-2 reported and recorded crime rose much faster than crime found in the British Crime Survey, suggesting exaggeration by police statistics of the increase over this period. Despite this, the broader trends shown are rather similar.
There is also a potential issue regarding the police’s lack of acceptance that an offence has taken place. It was estimated that they didn’t record 40% of crimes as they were categorised in the BCS. In some cases this means that they were recorded as different crimes, but others went completely unrecorded. This can be the case for a variety of reasons, among them being the belief that it was too trivial or that the incident hadn’t taken place, a lack of evidence, or the victim wishing the matter not be taken any further.
Victimisation studies such as the British Crime Survey are not without their drawbacks, Hazel Croall (1998) has pointed out a number of problems with them. Firstly, it is only possible to record crimes that victims are aware of, thus automatically excluding many. Additionally, the figures are reflective of the respondents’ definition and memory of events, which may deviate from the legal classifications. The survey is also restricted to households, meaning it omits crime committed against businesses or institutions. Furthermore, the earlier iterations of the BCS failed to interview those under 16, therefore excluding most crimes committed against younger age groups. Because of such issues it is important that studies like the BCS be treated with caution, however their inclusion of more crime than police statistics means that they are likely more reliable than those.
Overall, it has been estimated by Maguire (2002) that about 3% of crime committed in England and Wales results in a conviction. While some believe those actually convicted represent a cross section of those committing offences, it is also possible that some sections of society are much more likely to be convicted than others, regardless of whether they’ve committed more crime. One method of investigating this is the use of self-report studies, in which individuals provide information about themselves and the number of crimes they’ve committed. Upon reviewing 40 of these studies on delinquency, Stephen Box (1981) said “we should be very sceptical of those who continue to argue that delinquency is located at the bottom of the stratification system.” Graham and Bowling (1995) conducted a more recent study, finding that social class had no influence on whether British people would admit to having committed an offence. Although, it was found that members of lower classes were more likely to admit having committed more serious offences.
Self report studies do suggest a consistent police bias against working class delinquents. Chambliss’s (1973) study of two American delinquent gangs backs up this idea, finding that the police often don’t take middle-class delinquents seriously, because they don’t conform to the police’s images of typical delinquents. This meant that the middle-class “saints” he studied were able to convince authorities that their delinquent activities were simply harmless pranks. However, there is some question over the accuracy of self report studies, as those replying may not always be truthful. Various tests involving lie detectors, and the questioning of adolescents’ friends have been carried out to check on the results of such studies, ultimately having found that about 80% of respondents told the truth. While they may not be entirely reliable, their location of many more offenders than those convicted makes them largely more reliable than official police statistics.
To conclude, official statistics can be useful when looking at broader trends of crime, but overall it would seem that other methods of study are significantly more effective. Due to the vast amount of incidences of delinquency not found in the official statistics, it seems that other methods such as self-report and victimisation studies are much more useful ways of looking at criminal activity. Official statistics nevertheless remain useful when considering the effectiveness of the police and courts, and the disparity between unreported and recorded crime.