The onset of the delinquent career

View the most recent version.

The age of onset of a delinquent or criminal career refers, in principle, to the age at which the person first commits a delinquent or criminal act. Where, as in the present study, data are available for childhood offending, the notion of a "delinquent or criminal act" is problematic – both because of the legal doctrine of doli incapax,¹ and because most if not all criminal law regimes have a minimum age of criminal responsibility: in Canada, since 1984, it has been the 12^th birthday. Thus, the "age of onset" in this study refers to the age at which a person is first recorded as being involved in an incident in connection with which he or she was chargeable with a criminal offence, or would have been chargeable if he or she had been at least 12 years old. As mentioned previously, it is possible that the public under-reports crime by children, and that the police under-record apprehended offenders younger than 12 years old because they cannot be charged. Therefore, in this study, offenders with an age of onset under 12 may be under-represented. Of course, the reliance of this study on police data means that early offending which is not known to the police is not captured: thus, the age of onset is that of the police-recorded delinquent career, not necessarily the "actual" delinquent career.

The age of onset is interesting in itself, but it has also been found to be correlated with many other aspects of the delinquent or criminal career. Indeed, the distinction between the characteristics of "early onset" and "late onset" careers and offenders is one of the most firmly established in the literature.²

Figure 25 shows the ages of onset of offenders in each cohort. Very small numbers in both cohorts are recorded early onset offenders: less than 0.5% of each cohort commit their first recorded offence at each year of age before 11, and less than 1% at 11. Many more offenders begin their delinquent careers at 12 years or older: 1.5% of the 1990 cohort at 12 years, 2.5% at 13, and so on, with a peak of 3.7% of the 1987 cohort beginning their delinquent careers at the age of 15. After 15, the numbers of new offenders begin to decrease.

Recorded proportions of the 1987 cohort with ages of onset of 8, 10, and 12 to 14 are higher than in the cohort born in 1990.³ The reason for this is that there are no data about offending prior to 1995, and therefore no data on onset prior to their 8^th birthdays of members of the cohort born in 1987. Thus, some members of the 1987 cohort are recorded as beginning their careers at ages 8 to 14 (and older), who actually committed their first chargeable offence at 5 to 7 years of age.⁴ In view of this slight inaccuracy in age of onset statistics for the 1987 cohort, the approach used in the following analyses of age of onset (as in the analyses of cumulative prevalence of offending, above) is to combine the cohorts by reporting proportions of the cohort born in 1990 for the ages of 5 to 13, and of the 1987 cohort thereafter.

To recapitulate, the recorded onset of offending is very low during childhood and rises rapidly during the teenage years. Although "early onset" is a key concept in criminal careers research, it has no agreed-upon definition. If early onset is defined as occurring before the minimum age of criminal responsibility in Canada - the 12^th birthday – then only 1.7% of the 1990 cohort, or 11% of recorded offenders in both cohorts, were early onset offenders.⁵

Figure 26 shows the changes with age, relative to the level at the previous year of age, in the proportion of the combined cohorts committing their first recorded offence.⁶ Although the proportions increase with every year of age up to 15 (Figure 25), the relative change with age (Figure 26) shows no such pattern. Apart from the possible undercounting for all ages prior to the minimum age of criminal responsibility, which is reflected in the chart by a large increase at the age of 12, the change from year to year in the proportions of the cohorts who committed their first recorded offence fluctuates between the ages of 6 and 11, then decreases rapidly from 12 to 17 years. Relative onset becomes negative at 16 years of age – that is, the number of cohort members whose recorded careers begin after the age of 15 becomes smaller instead of larger with each year of age (cf. Figure 25).

---

Notes

1. Literally, "incapacity to do wrong". This is the English common law doctrine that children below a certain age (originally, 7 years old) are mentally incapable of forming criminal intent (Bala, 2003: 167-168).
2. The classic statement is by Moffitt (1993), but the distinction is employed and analyzed by many studies of delinquent and criminal careers.
3. Two-tailed t-values for the differences between the proportions of the two cohorts with ages of onset of 8, 9, ..., 14 are 2.75, -1.09, 1.69, -1.56, 2.06, 1.87, and 5.78. These statistics should be interpreted with caution, since these are not normally distributed random samples. Satterthwaite's (1946) approximation for the degrees of freedom was used, since the variances of each pair of vectors of observations were unequal (SAS Institute, 2004). The t-values for ages 8, 12, and 14  are significant at p<.05 (p₀₈=.006, p₁₂=.039, p₁₄<.0001). See note 1 in "The age-crime curve" for a discussion of the difference at the age of 14.
4. Or younger than 5 years, but that is very rare indeed.
5. If a "very early onset" offender is defined as one whose first recorded offence occurred before the 8^th birthday, then only 0.22% of the 1990 cohort, or one in every 460 cohort members, falls into this category. To put it differently, less than 1% of recorded offenders aged 5 to 17 had very early onset. If early onset is defined as occurring before the 14^th birthday, then 5.5% of cohort members, or 40% of recorded offenders in both cohorts, had early onset.
6. Defined as: (x_t - x_t-1)/x_t-1, where x_t is the proportion of the cohort recorded as committing their first offence at age t.