3.3.1 Official statistics and self-report studies
We will return to Muncie’s observation about ‘media-driven panics’ in the following section, but here we want to consider other patterns of social difference in relation to the control of juvenile delinquency. Using two different types of evidence, sociologists have pointed to a systematic difference between who breaks the law and who is officially labelled as law-breaking. On the one hand, official statistics about crime – data collected by the criminal justice system that records arrests, prosecutions, sentences, and so on – identifies delinquency and criminality as a behaviour associated with young working-class men. On the other hand, self-report studies – in which people report their own law-breaking activities anonymously – tend to contradict this apparent distribution. Muncie summarises the situation as follows:
… the major contribution of self-report studies has been to seriously question widely held beliefs about the correlations of class position, ‘race’ and gender to criminality. Both Anderson et al. (1994) and Graham and Bowling (1995) found that middle-class children were just as likely to be involved in crime as working-class children. Indeed a survey by the British Household Panel in 2001 based on interviews with 1,000 13−15 year olds found that those from higher income families were more likely to commit vandalism, play truant and take illegal drugs (Guardian, 25 February, 2001) … This suggests strongly that official statistics reflect not patterns of offending but patterns of policing. As a result, the relative criminality of certain groups of young people has been exaggerated. For example, inner-city working-class youths face a greater risk of arrest than middle-class youths engaged in similar activities but in areas where the police presence is lower. Ethnic minority youths are statistically more likely to be stopped and searched by the police (Burke, 1996), but self-report studies show that those of Indian, Pakistani and Bangladeshi origin have significantly lower rates of offending and that for African-Caribbeans the rate is no higher than for whites.
There are two important points to draw out of this discussion. The first concerns the problem of evidence in relation to delinquency and criminality. Two types of data – official statistics and self-report studies – produce very different pictures of the social distribution of law-breaking behaviour. Each of them has problems. As Muncie suggests, official statistics tend to report police activity rather than law-breaking. Self-report studies also have potential flaws, as people may be selective about what they admit to; or may even over-dramatise their law breaking when hidden behind anonymity. Such studies tend to be conducted through questionnaires with relatively low response rates; and tend to focus on a limited set of crime behaviours (leaving out a range of possible crimes – from large-scale financial fraud to child abuse). Evidence in the social sciences is rarely perfect, but here it is the comparison between the two sorts of data that is the point of interest.
The second important point concerns the shift of attention to the processes of social control, and their critical role in defining and acting on what – and who – counts as deviant. This concern with ‘defining the situation’ inspired a wide range of investigations into crime, deviance and social problems more generally. For example, Bacchi’s work (1999) builds this starting point into an analysis of social policies and their relationship to gender divisions, arguing that the power or capacity to define ‘what the problem is’ is central to the policy-making process. This issue is also central to other work on youthful disorder.