Reconviction over Time

As explained above, the prisoners in this sample had been out of prison for different periods of time and obviously the longer someone is out of prison the more opportunity there is for that person to be reconvicted. In order to get a truer picture of the proportion reconvicted as time from release from prison progressed, it was necessary to standardize follow-up periods, ignoring any reconvictions that occurred after each follow-up period. Naturally, the longer the follow-up, the smaller the sample became. We compared the reconvictions of all prisoners up to and including:

	2 years after their release, N = 173
	4 years after their release, N = 162
	6 years after their release, N = 94

Table 2 shows that the percentage of prisoners reconvicted of the different categories of offence increased as the follow-up period lengthened, but even so it remained comparatively low. Thus, the reconviction rate for a sexual offence leading to imprisonment increased from 1.1 per cent after two years, to 4.3 per cent after four years and to 8.5 per cent (1 in 12) after six years. For a sexual or violent offence resulting in imprisonment, the rate of reconviction almost trebled between a two-year and six-year follow-up from 4.6 to 12.8 per cent (1 in 8). Although 30.9 per cent of those who could be followed-up for six years had been reconvicted at some time, only one in six (18.1 per cent) had been yet again imprisoned.

Table 2
Type of reconvictions of sex offenders by ti me since release from prison

Type of reconviction	Two years		Four years		Six years
Type of reconviction	N	%	N	%	N	%
Any sexual offence and imprisoned	2	1.2	7	4.3	8	8.5
Violent offence and imprisoned	6	3.5	8	4.9	4	4.3
Total sexual or violent offence imprisoned	8	4.6	15	9.3	12	12.8
Other offence and imprisoned	4	2.3	6	3.7	5	5.3
Total imprisoned	12	6.9	21	13.0	17	18.1
Reconvicted but not imprisoned	10	5.8	17	10.5	12	12.8
Total reconvicted	22	12.7	38	23.5	29	30.9
Not reconvicted	151	87.3	124	76.5	65	69.1
Total sample	173		162		94

These figures are not cumulative because some prisoners included in one follow-up period are excluded from the next. Also, some prisoners in one category at one follow-up period moved to a higher category when the follow-up period became longer.

How do these findings compare with those of other studies?

Marshall (1994: 26) found that 7 per cent of 402 sex offenders released from both short- and long-term prison sentences in 1987 were reconvicted of a further sexual offence within four years of their release from prison. In a later study of the reconviction rates of sexual offenders in

[Page 379]

general (both imprisoned and not imprisoned) drawn from a number of birth cohorts, Marshall found that 10 per cent were reconvicted of a further sexual offence within five years of their first such conviction and that altogether 22 per cent were reconvicted of a sexual or violent offence within the same period, although he gave no indication of the seriousness of the offences (1997: 3).

Friendship and Thornton (2001) recently reviewed the literature and carried out their own study of more than 1,000 sex offenders released from a prison sentence of four years or more in 1992 and 1993. They found that 5 per cent were reconvicted of a sexual offence within four years of their discharge, a figure very similar to the finding of this study at the four-year period. However, Friendship and Thornton believed that the true reoffending rate was probably considerably higher. This is discussed below.

It is known that recidivism rates for sex offenders vary according to the type of victim (Centre for Sex Offender Management 2001). Thus, we compared the reconviction rates for those who had served their prison sentences for sex offences against adult victims (strangers or non-strangers) with those whose sex offences were against children (intra- familial or extra-familial). The findings, over both the four-year and the six-year follow-up periods (see Tables 3 and 4), show that:

	None of those who had originally offended solely against children in their own family unit was reconvicted and imprisoned for a sexual or a serious violent crime, even when the follow-up period was extended to six years.
	Thus, all the offenders against children who were reconvicted for a sexual offence had originally been convicted of an extra-familial sexual crime. And of the 19 who could be followed up for six years, all five who were convicted of a further sexual crime victimized another child. The numbers are small but the difference between intra- and extra-familial offenders is striking. By six years, 32 per cent of the extra-familial group had been imprisoned again for a sexual or serious violent crime.
	The reconviction rate for sexual crime of sex offenders against adults had, by six years, reached 7.5 per cent (3/40) , and for a sexual or serious violent crime 15 per cent ( 6/40) , whether the original offence had been committed against a stranger or not.

Table 3
Type of reconvictions of sex offenders by original offence type: four-year follow-up

Type of reconviction	Offenders against children				Offenders against adults				Total
	Intra-familial		Extra-familial		Non-stranger		Stranger		Total
	N	%	N	%	N	%	N	%	N	%
Any sexual offence and imprisoned	0	0	3	9.1	3	9.7	1	2.9	7	4.3
Any sexual or violent offence and imprisoned	0	0	5	15.2	4	12.9	6	17.6	15	9.3
Other offence and imprisoned	1	1.6	1	3.0	2	6.5	2	5.9	6	3.7
Total imprisoned	1	1.5	5	18.2	6	19.4	8	23.5	21	13.0
Reconvicted but non imprisoned	4	6.3	1	3.0	6	19.4	6	17.6	17	10.5
Total reconvicted	5	7.8	7	21.2	12	38.7	14	41.2	38	23.5
Not reconvicted	59	92.2	26	78.8	19	61.3	20	58.8	124	76.5
Total sample	64		33		31		34		162

Table 4
Type of reconvictions of sex offenders by original offence type: six-year follow-up

Type of reconviction	Offenders against children				Offenders against adults				Total
	Intra-familial		Extra-familial		Non-stranger		Stranger		Total
	N	%	N	%	N	%	N	%	N	%
Any sexual offence and imprisoned	0	0	5	26.3	2	9.5	1	5.3	8	8.5
Any sexual or violent offence and imprisoned	0	0	6	31.6	3	14.3	3	15.8	12	12.8
Other offence and imprisoned	1	2.9	1	5.3	1	4.8	2	10.5	5	5.3
Total imprisoned	1	2.9	7	36.8	4	19.1	5	26.3	17	18.1
Reconvicted but non imprisoned	3	8.6	2	10.5	4	19.1	3	15.8	12	12.8
Total reconvicted	4	11.4	9	47.4	8	38.1	8	42.1	29	30.9
Not reconvicted	31	88.6	10	52.6	13	61.9	11	57.9	65	69.1
Total sample	35		19		21		19		94

All but one of those reconvicted of a sexual offence offended against a victim in the same age category as the victim of the crime for which they had been imprisoned. Only one of the six offenders against an adult who were reconvicted for a sexual offence victimized a person under the age of 16. All five offenders against children who were reconvicted chose child victims.

To what extent was this evidence of specialization?

Soothill et al's long-term study of the criminal careers of more than 6,000 sex offenders revealed that, whatever other crimes they had committed, when it came to convictions for sexual offending they tended to be specialists rather than generalists: 'that is, if they were convicted of a sexual offence on another occasion, then it was most likely to be a conviction for the same type' (Soothill et al. 2000: 66).

To test whether this was true of this sample, the criminal careers of the offenders, both prior and subsequent to the current offence, were examined.

The analysis revealed that three-quarters (77%) of the 162 men who could be followed-up for four years had no prior and no subsequent convictions for sexual offences. But amongst those (37 of the 162) who had more than one conviction for a sex offence, there was evidence of a degree of 'specialization' in their reoffending patterns.

[Page 381]

Nevertheless, a sizeable minority of recidivist sexual offenders switched between child and adult victims and vice versa. Thus:

	Of 24 whose first sexual conviction was for an offence against a child, 71% repeated their sexual offences against children only, the rest switching to an adult victim on at least one occasion,
	Of 13 whose first sexual conviction was for an offence against an adult, 62% repeated their sexual offences against adults only, but 38% switched to a child victim on at least one occasion.