re: No charges filed against Jameis Winston

quote:

---

The median of all the cited studies on Wikipedia is 11.8%

---

It also failed to mention the McDowell data, which I think is the most reliable of all the studies because of its adequate sample size and straightforward methodology. And also because the data was subsequently confirmed by additional data that was forcefully swept under the rug for political reasons.

quote:

---

Dr. Charles P. McDowell of the U.S. Air Force Office of Special Investigations led a team that investigated 556 rape allegations. The initial research categorized 220 of those allegations as true, 80 as false, and 256 as inconclusive. When they announced their initial results, they'd categorized 46% of their data as inconclusive. This strongly suggests that their approach was cautious and objective. Advocacy researchers know the answer they want before they even start. It's unheard of for advocacy researchers to announce that the largest portion of their data was inconclusive.

The McDowell team then did a followup analysis. They recruited three independent reviewers to evaluate the inconclusive cases based on a list of criteria that were common among the women who had acknowledged they lied. (See below for an example of the sort of criteria that were used.) In order for any of the inconclusive cases to be recategorized as false, all three independent reviewers had to agree that it was false. The end result: 60% of the 556 rape allegations were determined to be false.

Even so, McDowell worried that his findings might not be representative outside a military context. So he then analyzed the police records of a major midwestern city and a city in the southwest. The results in those cities were consistent with the results of the Air Force analysis. But the cities feared political repercussions if it were publicly announced that 60% of reported rape allegations were false, so they requested anonymity.

---

Scientists and researchers came come under a lot of criticism for publishing data that might be politically unpopular.* Which is why it's important to look at the methodology (i.e. read the reports) yourself and figure out what is credible.

What I'm getting at, is if you actually think the number is 11%, you're being naive. And I will acknowledge that, for various reasons, the percentage of fabricated rape claims is a tough number to definitively pin down. But based on the information I've read on the subject -- and I've read a lot...I wrote a paper on it in a law school seminar class -- any claim that rape accusations are being fabricated at a rate of less than 1 in 3 fails the straight-face test.

_____________________
*See "The Bell Curve", which was, and is, attacked for doing nothing other than presenting data. And doing so in the most meticulous and comprehensive manner that data could ever be presented.