McMullin et al (2007) draw what they frame as a startling conclusion from their 3 year study of undergraduate women with or without a history of sexual victimization. They found that victims displayed less positive feminine personality traits and more negative masculine personality traits than non-victims. McMullin et al state: "This is contrary to
Muelenhard and Linton's (1987) suggestion that sexual victimization may lead to more traditional attitudes."
However, this does not appear to be the conclusion in Muelenhard and Linton's assessment of risk factors for date rape at all: "men who had engaged in [sexual assault] were more traditional than other men, whereas women who had experienced [sexual assault] were less traditional than other women."
Anderson and Lyons (2005) found that although men were more likely to "blame the victim" than women in cases of sexual assault, this difference was mediated by acceptance of traditional gender roles.
Howells et al (1984) also found that attitudes towards women was a good predictor of attitudes towards rape in both men and women.
I'm probably way off base here, but it doesn't surprise me that McMullin drew the same conclusion as Muelenhard and Linton. It seems unlikely to me that people who would be willing to discuss their sexual assault at three follow-up interviews would continue to blame themselves for the assault. While I wouldn't be surprised to see some reports of feelings of self-blaming or guilt from assault victims, I would expect to see far fewer reports of sexual assault among people who tended to blame themselves. It seems (to my completely unqualified eye) like this is a reporting question more than a causal one.