Testing the Assumption of Behavioural Consistency in a New Zealand Sample of Serial Rapists

One of the assumptions that underlies the profiling process is that criminals are behaviourally consistent from one offence to another. To date, however, this is an assumption that has not been scientifically validated. The present study therefore tested the assumption of behavioural consistency in serial rape offences. The author collected dichotomous data on 30 behavioural variables for a total of 439 offences committed by 121 serial rapists in New Zealand. There were two main research aims of the study. The first aim was to test the behavioural consistency of a range of individual behaviours. It was hypothesised that higher consistency would be found for behaviours that reflected a degree of planning or that prioritised control of the victim and the offence environment, because these behaviours might be less affected by environmental factors. In contrast, many sexual behaviours arise directly out of offender-victim interactions and therefore are most affected by environmental factors such as victim resistance. It was therefore also hypothesised that sexual behaviours would display lower consistency. A consistency measure was used that compared behaviour in consecutive offences. Consistency for each behaviour was defined as present-present or absent-absent matches of that behaviour in consecutive offences. The degree of consistency for any behaviour will be reflected in the consistency score received by that variable based on the number of matches for that behaviour across the offence series. The consistency analysis found moderate to high levels of consistency for the majority of individual behaviours. As predicted, higher consistency was exhibited for behaviours that prioritised control of the victim and the offence environment, and lower consistency was exhibited for the sexual behaviours. The second research aim was that if behavioural consistency was found in the results of the consistency analysis, to explore whether there were any underlying patterns to the consistency of offending behaviour. A factor analysis of the consistency scores established that there are clear patterns to the behavioural consistency of offenders consistent with previous analysis of offence characteristics. The factor analysis resulted in three themes or domains to behaviour: hostility, involvement and control. These findings have theoretical implications for the assumption of behavioural consistency in serial rapists, for the concepts of modus operandi and signature in offence behaviour, and for the theoretical understanding of the profiling process. The findings also have practical implications for the practice of profiling and case linkage in New Zealand, and raise possibilities for future directions in research