The present study investigated the attrition process of rape allegations. The study adopted a multi-method research approach incorporating quantitative and qualitative data to reveal the factors that influence the processing of rape and the degree to which stereotypical notions, or 'rape myths', can still be held responsible for the poor outcome of so many cases. Analysis revealed that the majority of rape allegations did not proceed beyond the police stage of proceedings, complainant withdrawals accounting for the majority of these. Just 21 per cent of the original sample reached the Crown Court, resulting in a final conviction rate of six per cent. The research established that the views of those involved in investigating and prosecuting rape cases were often based on stereotypical views of 'genuine' rape, usually rooted in notions of appropriate female behaviour. Thus, cases approximating the 'classic rape' - involving stranger perpetrators, virginal victims, the use - or threat - of a weapon, and evidence of injury to the complainant - were those most likely to proceed. However, in a climate where the nature of rape has changed to include more allegations involving complainants and perpetrators who are known to one another, physical evidence is often lacking and the issue becomes one of 'consent' and whether the complainant willingly engaged in sexual relations with the alleged perpetrator. In this case, the focus turns increasingly to the character and the credibility of the complainant and whether she will present as a strong witness in court. Indeed, what is found to be at the centre of the debate is not necessarily whether there is a 'genuine rape' or even whether there is a 'credible victim', but, more usefully for the purposes of the criminal justice system, whether there is a convincing witness to put before a jury - a consideration that was found to affect decision-making throughout the system.