False Rape Allegations Are Rare

Any regular readers here are likely to already know this. The reputable, methodologically sound reviews put the frequency of false reports in the single-digit percentages. There are people who, for propaganda reasons, keep saying that the incidence of false reports is much higher. They create these figures with biased reviews or intellectually dishonest mislabeling. In an effort to add more sunlight to kill those germs, I’m going to give you tomorrow’s headlines today. A new study is set for publication this December in the journal Violence Against Women, based on a review of every single rape allegation made to a US university police department — the study does not disclose which school — over a ten year period. The result: 5.9 percent false allegations.

The lead author is David Lisak, a psychology professor at the University of Massachusetts who specializes in interpersonal violence. He is one of the few psychologists to study rapes committed by assailants who are not institutionalized, and I have written about his work many times before. See here, here, here, paper online here, and a summary here.

Lisak’s methodology was simplicity itself. Four researchers paired up into two teams. Each team went through the entire universe of rape reports, having access to the whole file. They used four categories:

False Report: After a thorough investigation, evidence showed that the assault had not occurred.
Case Did Not Proceed: Whether because of insufficient evidence, inability to identify the perpetrator, the survivor withdrew from the process or the survivor’s account did not meet the definition of a sexual assault.
Case Proceeded: Either formal or unformal disciplinary measures were taken.
Insufficient Information: The file lacked basic information necessary to categorize it.

After the two teams independently reviewed all the files, the compared coding to see if they all agreed on the coding. There were no disagreements on the false report cases and only seven overall.

One important part of the paper is the definition they used for false reports. They didn’t make it up. They applied the guidelines issued by that notorious bastion of feminist indoctrination The International Association Of Chiefs of Police:

The determination that a report of sexual assault is false can be made only if the
evidence establishes that no crime was committed or attempted. This
determination can be made only after a thorough investigation. This should
not be confused with an investigation that fails to prove a sexual assault occurred.
In that case the investigation would be labeled unsubstantiated. The
determination that a report is false must be supported by evidence that the
assault did not happen.

Applying the IACP guidelines, of 136 cases reported to the university in ten years, eight, or 5.9%, were false reports.  (44% resulted in no disciplinary action.  35.3% resulted in the case proceeding, and the file in about 14% was too insufficient to code.)  It is perhaps important to note that the coders agreed with the university police about the eight false report cases:  There were only eight files that the university had deemed false reports, each coding team indepently agreed with those determinations, and neither coding team found a file that should have been deemed a false report but was not.  So essentially over a ten year universe of cases three independent groups of people reached precisely the same conclusions about which reports were false, and it was less than six percent.

In three of those eight, the complaining witness admitted fabrication during the investigation.  A fourth involved a partial admission of fabrication with supporting evidence of fabrication.  In three cases, careful investigation revealed that the complainants’ story was likely untrue even though the complainant stuck to the story.  The last of the eight Lisak’s paper describes this way:  “The complainant recanted her report but the facts yielded by the investigation suggested that her initial report was as much a mislabeling of the incident as a deliberate effort to fabricate.” So in the one instance that could be called ambiguous, each team of coders independently decided to make the judgment call in the direction that would produce the higher number of false allegations; one could apparently argue that the number should be seven and not eight.

Lisak’s paper included a literature review that explains a great deal about how people who want to artifically inflate the number of false reports do so.  They simply sweep in other categories!  If “false report” is a catch-all category that included not only actual false reports but cases that the police decide not to prosecute because the victim will not be a good witness, or does not want to press charges, or cases that are true but where they can’t find the assailant, or cases where the conduct the survivor describes doesn’t make out the elements of the offense … well, a single-digit percentage can quickly swell to eye-popping and propaganda-friendly numbers.  That kind of thinking would lead a researcher to code this as a false allegation, because the cop didn’t want to take a rape report from a women who had been drinking, even though the subsequent evidence in the civil case, including sworn deposition testimony, make it pretty clear that it did happen. Or, a researcher could say that this case was a false report because the jury acquitted, though the rapist admitted it happened and the jury of assholes decided that sex workers are outside the reach of legal protection.

This kind of sweeping-in is easy to do when the researcher does not have access to or does not look at the underlying facts.  If a police department only prosecutes half the rape complaints, and a researcher is of a mind to deem as many cases false as possible, he can simply call the other half false.  That’s not methodologically sound or even intellectually defensible, but it’s a easy way to produce a big number.  Which, in turn, gives people excuses to act like skeptics and inquisitors to women who report allegations of sexual assault.  Which it turn causes more women to decline to report.

False reports sometimes happen.  False reports are a single-digit percentage of the total reports, which is already a massive undercount of all sexual assaults because the majority of rape survivors decline to report.  The incidence of false reporting is simply not high enough to justify the propaganda put forth by the pro-rape lobby.