Mark Duckenfield writes:
Some comments on statistics and “bad math”, that I think display a clear misunderstanding of statistics and surveys.
and the editorial to which it refers
The original report is quite clear about weighting things, smaple sizes, etc. The apparent “clincher” argument in the editorial—that over 50% of unwanted sexual advances in the military are experienced by men—seems to confound several things. The first being, of course, that with only 14% of service members being female, it is quite likely by their broad definition that the 86% of service members who are men actually *do* comprise the majority of unwanted sexual attention even if the incidence is only a fraction of that of female service members. Of course, the general population, which is over 50% female would have well over 50% of sexual advances experienced by women. Which is not to say that there might not be problems of men reporting male-on-male improprieties—given that the first response tends to be either “it was consensual” or “they wanted it”, I wonder how many heterosexual men in the armed forces are likely to bring a complaint in such circumstances? Especially since “unwanted sexual advance” is a rather broad definition.
I haven’t looked into this enough to have any informed reaction, just wanted to share with you as this indicates the difficulties of measurement and the challenges of interpreting measurements.