Special Problems with Analysis of Serious Crimes of Violence

Violent crime accounts for approximately 7% of recorded crime statistics, and of this only a very small number of crimes fall into the category of being very serious, such as extreme violence or rape. This means that in the BCS sweeps there are insufficient serious cases for any meaningful analysis at this stage, although the larger samples sizes in future years may assist in collecting sufficient cases for analysis.

The available data on serious violent crimes have been of a poor quality. A number of factors compromise the quality of such data, including:

• the accuracy of recall from victim statements, especially if they have been the victim of a particularly traumatic crime (e.g. Grubin, Kelly and Bransdon, 2000);

• the reliability of eyewitness statements (e.g. Kebbell and Wagstaff, 1999);

• variations in both interviewing techniques and statement-taking across police forces will affect the amount and quality of information obtained concerning a crime (Clarke and Milne, 2001);

• the loss of the primary data source in homicides as there is no victim available to question;

• identifying what data should be collected has proved problematic as there are often conflicting requirements and low levels of compliance by forces in providing the data;

• variation can also come from the large numbers of people required in data collection and coding;

• inconsistent coding;

• missing information;

• updating and maintaining the data collected.

This problem is particularly acute in relation to homicide. In the United Kingdom most of what we know about the characteristics of homicide is taken from the Home Office Homicide Index (HI). A similar dataset on homicides, in Scotland, is maintained by the Scottish Executive. The Homicide Index is primarily an administrative database that collects details of incidents initially recorded as homicide by the police. The Index was started in 1967, with modifications made to the scope of the information in both 1977 and 1995 and is maintained by the Home Office's Research, Development and Statistics Department.

The Homicide Index may offer an important contribution to assisting the investigation of hard-to-solve homicides by allowing investigators to consider the characteristics of an individual case against detected cases with similar characteristics. The use of national datasets is not new. The database known as CATCHEM, which is maintained by the Derbyshire Constabulary, contains the records of child homicide victims from 1960 to date (Aitken et al., 1995). The database known as BADMAN, which is maintained by the Surrey Constabulary, provides support in respect of child homicide offenders and stranger rapists. The National Crime Faculty has been established at Bramshill, the United Kingdom's national police college, to provide an analytical capability to deal with serious violent crimes but these data are not published. The Serious Crime Analysis Section (SCAS) conducts comparative case analysis on murder, rape and abduction cases that fall within certain criteria. It is based on the Violent Crime Analysis System (ViCLAS) developed by the Canadian government and stores details of relevant offences covering 126 variables. Use of these databases can assist in identifying lines of enquiry and reduce the time taken to investigate serious crime. Empirical data sets have an important role to play in the investigation and analysis of serious crime but they are dependent on police forces supplying the relevant information and this has required the creation of rigorous quality control processes. Serious crime databases can identify the likely geographical relationship between an offence and an offender (Davies and Dale, 1996), the relationship between an offence and the conviction history of an offender (Davies, Wittebrod and Jackson, 1998) and the probable characteristics of the offender for given crime types (Aitken et al., 1995). These databases provide a good example of the use that can be made of crime data if there are sufficient controls over the quality of the data. See the chapter, in this Handbook, by Canter for illustration of the importance of quality crime data, including geographic, for effective crime investigation.

0 0

Post a comment