Age, Sex and Crime

In one line

Offending rises sharply in adolescence, peaks in the late teens or early twenties and falls away thereafter, and men offend several times more often than women, and no explanation of crime that cannot account for these two facts is worth much.

In the wording a student can write in an exam: the age-crime curve is the most robust finding in criminology: offending rises steeply from early adolescence, peaks between about seventeen and twenty-two depending on the offence, and declines throughout adulthood; and the sex ratio in offending is heavily male, most sharply for violence and least for offences such as fraud committed within the household or workplace.

The age-crime curve

The shape. Steep rise, sharp peak, long decline. It has been found in every country and every period in which it has been looked for, in official statistics and in self-report data alike, and it was already visible to Quetelet in the 1830s.

Property offences peak earlier than violent offences, typically in the middle to late teens for theft and in the early twenties for serious violence, and offences requiring position and access, such as fraud and corruption, peak far later.

Two competing explanations, and the examiner wants both.

The curve reflects a change in offenders. On this view most young offenders are adolescence-limited: they offend for a few years and stop, and a small group of life-course persistent offenders continues. The evidence for this is the Gluecks' follow-up and the modern cohort studies.

The curve reflects a change in circumstances. On this view the same people become less likely to offend as employment, marriage and children arrive, which is control theory's account: commitment, attachment and involvement all increase together in the twenties.

The two are not exclusive and the best supported position combines them: most desist because their circumstances change, and a small minority whose difficulties began early do not.

The Indian pattern is the same. Recorded offending in India is heavily concentrated in the eighteen to thirty band, and the concentration is sharper for the offences that require physical capacity.

Sex and crime

The finding. Men are recorded as offenders several times more often than women, and the ratio is widest for robbery, homicide and other violence and narrowest for offences such as theft within a household, some kinds of fraud, and offences relating to children.

Explanations, and their weaknesses.

The chivalry hypothesis. Women are less often reported, arrested, charged and convicted, because complainants, police and courts treat them more leniently, so the difference is partly in the recording. There is some support for this, and it does not survive as a complete explanation: the sex difference appears in self-report data too, though smaller.