Citation

McMillan D, Hastings RP, Salter DC, Skuse DH. Arch. Sex. Behav. 2008; 37(6): 877-890.

Affiliation

Academic Unit of Psychiatry and Behavioural Sciences, University of Leeds, 15 Hyde Terrace, Leeds, West Yorkshire, LS2 9LT, England, UK. d.mcmillan@leeds.ac.uk

Copyright

(Copyright © 2008, Holtzbrinck Springer Nature Publishing Group)

DOI

10.1007/s10508-007-9193-0

PMID

17653841

Abstract

Research on developmental risk factors for sexual offending against children is inherently difficult. Although there are numerous studies in the area, methodological problems are often not given sufficient attention in the design and interpretation of research. Despite the large number of variables retrospectively associated with sexual offending, the status of these as developmental risk factors remains unclear. If sexual offending research is to make progress, five methodological challenges must be met. These include: (1) establishing the temporal relationship between a putative risk factor and sexual perpetration; (2) ruling out third variable effects; (3) avoiding retrospective reporting bias; (4) generating informative tests of risk mechanisms; and (5) modeling complex developmental phenomena. In this review, we outline these methodological problems and suggest that the study of sexualized behavior in childhood and adolescence may provide some leverage on each of these difficulties.

Language: en