Pathways Of Delinquent Development

One of the best-proven criminological results is the 'age curve' of delinquent behaviour (Blumstein, Farrington and Moitra, 1985; Loeber, Farrington and Waschbusch, 1998). In relation to the population as a whole, young people show a disproportionately much greater level of delinquency. The incidence and prevalence rates of offending rise strongly in early adolescence and peak between ages 16 and 20 (depending on the respective kind of crimes). From early adulthood onwards, the rates of delinquency decreases. This characteristic age curve is not only found in official crime statistics (e.g. police data) but also in self-reports. Although males show much more delinquent behaviour, and in particular more violent crime, the age curve is similar for both sexes. In accordance with earlier maturation, however, female delinquency seems to increase and decrease a little earlier than male delinquency (Stattin and Magnusson, 1996).

The increase of offending in early adolescence is primarily due to youngsters who are registered only once (Farrington, 1992; Wolfgang, Figlio and Sellin, 1972). These are supplemented by a group that offend repeatedly but desist soon from delinquency. Both groups represent an adolescence-limited pathway of delinquent development (Moffitt, 1993). Approximately one-third of all young males become officially registered as having committed a crime. And that must be an underestimate, because many other crimes and individuals are not detected. So this kind of delinquency can be interpreted as a more or less normal transition of youth. Most typical offences are shoplifting, bicycle theft, and other petty property offences. Even cases of violent crime are often not very serious (e.g. robbery of baseball caps from other youngsters; fighting among rival groups).

In contrast to this adolescence-limited delinquency, a small group of 5-8% of young males continue offending into adulthood. It has been found that many of these offenders already exhibited aggression, delinquency, and other conduct problems during childhood (e.g. Farrington and Loeber, 2001; Patterson et al., 1998). Youngsters who follow this early starting, and relatively persistent, pathway of delinquent development are clearly over-represented among serious and violent offenders (Snyder, 2001). In late adolescence and young adulthood, more than half of the offences in each age cohort are due to this group (Loeber et al., 1998; Wolfgang et al., 1972).

Naturally, Moffitt's (1993) differentiation between adolescence-limited and life-course persistent antisocial behaviour does not cover the whole range of delinquent developments in real life. For example, Nagin and Land (1993) found three subgroups in the Cambridge Study on Delinquent Development.

1. Adolescence-limited. These have a maximum of delinquency around the age of 16 and mostly no convictions after age 21.

2. Low-level chronics. These demonstrate a slowly increasing registration until age 18 and relatively constant recidivism on a low level.

3. High-level chronics. These individuals develop a steep increase until age 18 and only a slow decrease in adulthood.

Other models of antisocial development differentiate between specific kinds of problem behavior (Loeber and Stouthamer-Loeber, 1998; Nagin and Tremblay, 1999). Loeber and Hay (1994), for example, suggest three developmental pathways from childhood through adolescence (see Figure 2.7.1).

1. Direct aggressive behaviour, such as bullying, hitting, fighting, cruelty to animals and later assault or rape (overt antisociality).

2. More indirect forms of antisocial behaviour such as shoplifting, frequent lying, vandalism, fire setting and later burglary, fraud, or serious theft (covert antiso-ciality).

3. Stubborn behaviour, defiance, disobedience, and later truancy, running away, or staying out late (authority conflict).

In all three pathways, the proportion of youngsters exhibiting such problem behaviour decreases with age, whereas the severity increases. There are young people who exhibit problem behaviour in all three areas (versatile antisociality). In cases of early starting, this group is equivalent to Moffitt's (1993) description of life-course persistent antisociality. Similarly, Patterson et al. (1998) describe a regular sequence of strong disobedience, anger outbursts, fighting, and stealing in adolescents that became 'chronic offenders' at age 18 years.

Although there is strong empirical support for the early-starting and long-term persistent pathway of delinquent development, one should not overestimate the continuity

Figure 2.7.1 Developmental pathways in disruptive/delinquent behavior Source: Loeber et al., 1999

of deviant behaviour. For example, Lahey et al. (1995) report that approximately 50% of boys with a diagnosis of conduct disorder did not remain in this category continuously over four years. This is a typical degree of problem stability from pre-school to school age (Campbell, 1995; Lavigne et al., 1998). Similarly, about one half of children with conduct disorders or extreme antisocial behaviour in childhood did not go on to serious criminal behaviour in adolescence (e.g. Moffitt et al., 1996; Robins, 1978). Although, as mentioned, Patterson and colleagues (1998) found a clearly persistent pathway, approximately half of the children who ranked high in antisocial behaviour at age 9 or 10 did not progress to early arrest and chronic offending by age 18. In the more specific area of aggression, there is also a large part of children whose problem behaviour is not stable over time (e.g. Haapasalo and Tremblay, 1994; Nagin andTremblay, 1999).

At first glance, this seems to contradict the relatively high stability coefficients for aggressiveness as found in meta-analyses (Olweus, 1979, 1994; Zumkley, 1994). Whereas average correlations are approximately 0.70 after one year, they decline as a function of the time interval between two measurement times. One should also bear in mind that such correlations only indicate the similarity of ranking orders between individuals and not the stability of the behaviour itself (Farrington, 1990,

2002; Loeber and Stouthamer-Loeber, 1998). Furthermore, assessors' biases in single informants also seem to contribute to high stability coefficients (Losel and Schmucker, in press).

Such arguments do not devaluate the theoretical and practical fruitfulness of a relatively persistent pathway of delinquent development from childhood through adolescence. However, emphasizing both continuity and change as basic principles of human development sets a realistic framework for the accuracy of long-term predictions in delinquent development. On the one hand, with more than 80% correct predictions such prognoses can be highly relevant for practice (e.g. Hawkins et al., 1998; Lipsey and Derzon, 1998; Losel, 2002). On the other hand, depending on the respective base rates and selection rates, there remain substantial proportions of false positives and negatives that must be addressed by differentiated explanations of the natural history of delinquent development (Losel, 2002).

0 0

Post a comment