CHAPTER EIGHT:

SEXUAL PREFERENCES OF VIOLENT SEXUAL OFFENDERS

PHIL WILLMOT AND CEDRIC HART

The issue of whether penile plethysmography can make a useful contribution to the assessment of rapists has generated considerable debate. In fact the debate can be divided into two questions:

A recent meta-analysis by Lalumiere and Quinsey (1994) summarising the results of 12 studies has seemed to demonstrate that convicted rapists do, on average, show different sexual preferences, to non-offenders-rapists as a group responded more to rape cues than non-sex offenders. There is also some evidence that rapists who show more deviant sexual preferences are more likely to re-offend (Rice, Harris and Quinsey, 1990).

The present study also seeks to examine this second question. The approach taken is to classify convicted sex offenders on the basis of their past record and then see whether those with a record suggesting that they are more inclined to sexual violence differ in their PPG responses.

METHOD

Sample

99 Cases from the sample described in the previous chapter for whom conviction history data were available. Nearly half of the subjects (47%) were serving Life Sentences and a similar proportion had been sentenced following a conviction for rape. About half the subjects had victims under the age of 14.

Stimuli and Experimental Conditions

These were as described in the previous chapter. The focus here is on response to the short video recordings depicting rape, non-sexual assault, or consenting sex. The effect of experimental condition will not be discussed here as it did not seem to moderate the validity of response to the rape preference material. However, there was a trend for more deviant responses to be produced to the rape material when it followed the age-preference material.

Analysis

All responses were transformed to Z-scores to remove the effect of the absolute level of response and then the Z-score to the normal (consenting) category was subtracted from the largest Z-score to a deviant (rape or assault) category to produce an index of rape preference.

Classification of Offence Histories

The offenders were broken down into four levels of risk for future rape and violence using the actuarial algorithm developed by Thornton and Travers (1991). Scores on this algorithm depend primarily on the number of convictions for rape and non-sexual violence.

RESULTS

Table 1
shows the mean rape-preference index for each of the four levels of risk distinguished by the algorithm  
 

Risk of future Violence

Mean rape-preference index

N or cases

Low

Medium low

Medium high

High

The amount of rape-preference systematically increases across the four risk categories with men who are at a high risk for future rape and violence showing the strongest rape preference. This trend is statistically significant (linear trend p<0.005).

CONCLUSION

Men with a record indicating a more significant risk of sexual violence tended to show a rape preference. These results support the idea that sexual preferences play a part in recidivistic sexual violence and suggest that PPG defined rape preference can make useful distinctions amongst convicted sex offenders.

REFERENCES

Lalumiere, M.L., & Quinsey, V.L. (1994) The discriminabitity of rapists from non-rapists using phallometric measures: a meta-analysis 'Criminal Justice and Behaviour,' 21, 150-175.

Rice, M.E., Harris, G.T., & Quinsey, V.L. (1990) A follow-up of rapists assessed in a maximum security psychiatric facility. 'Journal of Interpersonal Violence, 5, 435-445

Thornton, D. & Travers, R. (1991) A longitudinal study of the criminal behaviour of convicted sex offenders. In 'Proceedings of the Prison Psychologists Conference'. London: H M Prison Service.