Talk:False accusation of rape/Archive 2

Misplacement of "Estimates of the prevalence of false accusations"
The placement of this two sentence paragraph, the second and third sentences of the article, is absurd. Why would we bring up the issue of the possible over counting of false accusations before any information about the amount or frequency of such accusations is even presented? B-A-A-A-D!! Badmintonhist (talk) 18:53, 21 July 2013 (UTC)
 * Okay, then what shall we call the new section that your suggestion would require us to create after the sections on individual studies? –Roscelese (talk &sdot; contribs) 20:23, 21 July 2013 (UTC)
 * The subsections themselves give the reader plenty of notice that certain studies have been criticized for a methodology that might lead to over-counting false accusations; particularly the Kanin study. We actually don't need anything more than this. If there are reliable sources that comment on these studies collectively then we could place them in a concluding subsection. I must say, however, that it is rather difficult for me to picture a reliable source saying that all of these studies tend to overestimate the percentage of false accusations since the results are so disparate, ranging from the very low single digits to Stewart's quite incredible sounding 90 percent. Badmintonhist (talk) 00:10, 22 Starting off the sectionJuly 2013 (UTC)
 * As I've already explained, it's not our business to guess which studies the source is referring to and it would be quite inappropriate to put it in a section about any given study. Kanin's study has been individually criticized so that criticism is in the right section. If somehow you know which studies DiCanio is saying improperly overestimated false accusations, do enlighten us all. –Roscelese (talk &sdot; contribs) 03:58, 22 July 2013 (UTC)
 * Yes, "it's not our business to guess which studies the source is referring to" which means, of course, that we have no business placing that particular statement into our article. "I don't know which studies Kanin was talking about here but I'm gonna put this statement into the article anyway" makes absolutely no sense. Editors are supposed to make editorial judgments. Badmintonhist (talk) 05:41, 22 July 2013 (UTC)
 * Sounds like a pretty big IDONTLIKEIT to me. It's reliably sourced and explains the way in which such incorrect estimations come about. (The DiCanio source mentions FBI statistics, but, as we point out in the article, the FBI apparently doesn't distinguish between "unfounded" and "false," which isn't quite the same thing as saying that 8% are false.) –Roscelese (talk &sdot; contribs) 05:51, 22 July 2013 (UTC)


 * Actually it's an I JUST LIKE IT on your part. We are not obliged to include statements from reliable sources just because you like their political connotations.  You have already conceded that we don't know if Kanin was referring to all, most, some, or any of the specific studies that we have included in this section yet you still want to make his statement its centerpiece. Finding statements seem to support one's own views on a subject then trying ride them far beyond what they are worth is bad editing. Badmintonhist (talk) 07:17, 22 July 2013 (UTC)


 * I've already responded to the concern that the statement seemed to be a centerpiece or introduction, and based on your repeated misattribution, I doubt that you have actually looked more than superficially at the article or the sources for the sake of improving content, as opposed to seeing a chance to bother me and jumping in. Please stop bringing your personal grudges to article after article. If you're not going to make the basic effort necessary to have a conversation about making the article better - that is, reading the article/sources and reading the thread you're commenting in - I don't see that there is a point in continuing this discussion. –Roscelese (talk &sdot; contribs) 14:33, 22 July 2013 (UTC)

Nat'l Registry of Exonerations
Okay, it looks like there's potentially useful information there, although I had to go past the source that Judah cited (assuming good faith: you simply aren't aware that SAVE is a pro-false-rape group that lobbies against anti-violence legislation) and into the registry document myself. Here's what I started to write before I looked further:

"According to the National Registry of Exonerations, among 203 cases in which a conviction for sexual assault was found to have been wrongful, 23% had perjury or false accusation as a contributing factor."

This doesn't misleadingly state that "a significant" number were for perjury/FA when in fact that's just the opposite of what the source states (it points out that the leading issues were mistaken identification and bad forensics), and it also states clearly, while providing the exact number, that this is not "23% of accusations are false" but that we are dealing with a different set of data.

Thennnnnn I realized "wait, that sounds like a vague category, isn't it possible that we're adding in other information here" and continued to read the source. The number gets broken down into lies by supposed eyewitnesses (16%) and lies in cases without eyewitness deception (10%); the first number is further broken down into deliberate misidentification (8%) and fabricated crime (8%).

To sum up: Definitely a usable source. But not in the misleading and agenda-driven way that JudahH is using it, and possibly elsewhere in the article. –Roscelese (talk &sdot; contribs) 03:04, 23 July 2013 (UTC)
 * I clearly quoted enough text to make it clear what the 23% referred to. I never said anything like "23% of accusations are false". Obviously "significant" is subjective, but it's not the same as "majority of". I consider 23% a significant number, but in any case, I gave the exact number in the footnote.
 * The reason I didn't put numbers into the text of the article was because it was intended as a vague, general statement about ways in which the outcome of a case may not reflect the truth or falseness of the allegation, to balance out the original statement. If you don't think that this should be included in the article without placing it into detailed context, then surely neither should the statement about "many 'unfounded' cases are not false accusations". And again, if the information about wrongful convictions should go elsewhere in the article, then why should the information about 'unfounded' cases come at the beginning of the section? JudahH (talk) 03:22, 23 July 2013 (UTC)
 * You're misunderstanding. The sentence above was my starting to rewrite what you had added because you had used synthesis and unsuitable sources; I simply preferred clarity, since I was rewriting anyway. For the rest, I refer you to my repeated earlier statements about representing sources accurately without making up our own analysis and about possibly moving the material; no reason to waste my metaphorical breath. –Roscelese (talk &sdot; contribs) 03:33, 23 July 2013 (UTC)

Ahhh, and I've just figured out how to articulate the other problem with that edit. The DiCanio source literally and explicitly states that statistics on false accusation are complicated by unfoundedness. The Registry source does not state that statistics on false accusation are complicated by convictions turning out to be wrongful. We can use the Registry source - it's reliable, and the small number of wrongful convictions which are related to false accusation seems to be a usable statistic - but our WP:NOR policy prevents us from twisting the source to get a conclusion that isn't there, as JudahH is trying to do. –Roscelese (talk &sdot; contribs) 03:38, 23 July 2013 (UTC)

Definition
Please remember that Wikipedia reflects the definition used by the wider world, it does not dictate it. Since the bulk of the literature does not use "false accusation" to mean incorrect identification, police error, failure of forensic science, a real rape that didn't get a conviction, etc. it doesn't matter if a user personally thinks that the literature should - we use reliable sources here, and reliable sources use the term to mean an accusation by the supposed victim of an assault that did not occur. Even a really cursory search bears this out. –Roscelese (talk &sdot; contribs) 16:17, 17 June 2013 (UTC)


 * What about a situation where a rape has occurred but the victim has intentionally accused the wrong person? Relationships can get very complicated and I'm sure this has occasionally happened. Such an accusation would obviously also be false. It is also the kind of situation where forensic could disprove the accusation. Badmintonhist (talk) 18:27, 23 July 2013 (UTC)


 * Wikipedia isn't about situations that "could" happen or what words "could" mean, but rather about what reliable sources document. –Roscelese (talk &sdot; contribs) 18:41, 23 July 2013 (UTC)
 * In that case, it might be worth citing the reliable sources that give this definition in a footnote, at the least, wouldn't you agree? In point of fact, it doesn't seem to me that this definition is so cut-and-dried; rather, the various sources given in the article appear to vary on the point of definition, as Rhododendrites points out on this Talk Page. For example, there's already been a discussion on this Talk Page about the inclusion of the word "force" in the definition.


 * To the current point in particular, the following quotation from this article itself shows that not everyone uses the definition given in the article's first sentence:


 * Writing in the Law Review of Loyola of Los Angeles, Greer writes:
 * "Despite the difficulties in measuring wrongful accusations, there is indirect data available that is highly suggestive that far more than two percent of rape accusations are false. In a significant fraction of instances, the accusers recant their charges; in others, where no formal recantation occurs but where rape may have occurred, there are good reasons to believe that the accusation must nevertheless be wrong about the identity of the assailant. One illustration of this phenomenon are the instances where DNA testing has determined that the man actually imprisoned for rape after trial was not the individual the victim claimed was the assailant."


 * It would be useful for an introductory section to discuss the differences among the various studies, in terms of methodology, assumptions, and definition. I propose creating a section on this Talk Page to collate these differences before adding them to the article. JudahH (talk) 17:42, 24 July 2013 (UTC)
 * I've already clarified for Rhododendrites that the difference is just in methods, not in definition; it would be helpful if you would read the discussions that you are referring to. I see that Greer apparently uses a fringe definition, which (if we talk about him elsewhere) may be worth mentioning as context for his numbers, but which is inappropriate to elevate over the mainstream definition in the lede. (I'm also not sure that it's right to classify Greer as a study; his article rather seems to be an attack on other studies because he thinks they represent evil feminism.) –Roscelese (talk &sdot; contribs) 19:11, 24 July 2013 (UTC)
 * I did read the discussions that I referred to, thank you very much. It would be helpful if you would avoid snide comments. I know what you've said about definitions (everyone who matters uses the term as you've defined it), but you haven't backed that up with any evidence, and I just gave you a counter-example of someone cited in the article who does not use it in that way. Why don't you list a few of the sources that give your definition?


 * I'm not trying to attack you, by the way. Perhaps you're right that Greer uses a fringe definition, in which case that should be noted. But we should argue from the evidence, otherwise it's hard for a discussion to make progress.


 * Incidentally, just to clarify what we disagree on: it's the presumably-rare case where someone is in fact raped but deliberately accuses an innocent party of the crime. You take the position that this is not considered a "false accusation of rape", correct? JudahH (talk) 20:54, 24 July 2013 (UTC)


 * I ran a search in reliable sources and they use it in the same way. The only sources that throw in mistaken identification and so on seem to be anti-rape-prevention activist websites. If you seriously think it's necessary, we could easily pull in a bunch of sources (I think there was one cited for a bit) - I'd questioned the necessity of adding a whole section on definition when it's not really contested, but if there really is the potential for people misunderstanding the definition (instead of trying to use Wikipedia to change the definition) then we could do it. As for deliberate accusation of the wrong person, I'd want to look into more sources for it; it doesn't tend to appear in definitions, but that could either be because it's too rare to mention or because it doesn't count. The gist of my comment on the statistics, below, was that I wasn't sure (but that certainly some of the 23% figure isn't). –Roscelese (talk &sdot; contribs) 21:33, 24 July 2013 (UTC)
 * OK, I agree it's reasonable to look into the sources a little more to see exactly how they're applying the term. The most important thing, as I see it, is to ensure that it's clear how the expression is being used by various sources on the topic, especially those the article quotes. JudahH (talk) 05:04, 25 July 2013 (UTC)


 * You have provided zero proof that the unsourced narrow definition of the term that is currently used in the article is universally accepted among the reliable sources, or that the definition JudahH quoted is fringe.


 * In fact these well sourced wikipedia pages on false accusations and  that use four different sources to define the term (with half of the the sources dealing with the topic of sexual assault and rape as shown by their title, and other half possibly dealing with rape in the context they are referenced) directly oppose your claim.


 * They define the term false accusation to include intentional and unintentional lying of the accuser, misidentification, deliberate or accidental suggestive questioning, and faulty interviewing techniques. In fact the only time those well sourced pages even mention the narrow definition you claim is universally accepted, for which you provided no proof, was in the fourth source in the context of proposing a change in terminology. The fact that they are proposing a change in terminology to match your definition strongly suggests that you are incorrect, and that this is not how the term is generally, let alone universally, defined. If it was there would be no need to propose changes to the definition.


 * The current definition used in the article is unsourced because it is basically a modified version of the definition that Kanin study used. As someone who has been following this article for a while, but never participated, I have traced that the definition of the term was originally changed at this point . Then that changed definition was used to justify both the past (several days prior) and future (until present) removals of very notable studies that show how 23 and 26 percent of the primary rape suspects were exonerated using DNA evidence.


 * To sum it up:


 * 1) The definition of false accusation on this page is unsourced, and is far narrower than the well sourced definitions of false accusations on related pages.


 * 2) All four sources on other wikipedia's false accusation pages, as well as JudahH's source, disagree with your claim that the narrow definition of the term is universally accepted.


 * 3) Including all types of false rape accusations would make this article more informative and useful, and it would make related wikipedia articles more consistent. Confusion can easily be avoided by clearly defining what the term means. I suggest copying the definition from the previously mentioned wikipedia pages on false accusations.


 * 4) In the process of investigating why the well sourced DNA studies have been repeatedly removed I formed an opinion that you have repeatedly engaged in edit wars, tried to monopolize this article, as well as repeatedly violated good faith assumption by unfairly accusing other editors of trying to push their agenda and boost the rate of false rape accusations  This is not only my opinion, you were already reported for your behavior in this article . Contrary to your accussations, it is my opinion that you have been trying to do the opposite - to minimize the rate of false rape accusation and to repeatedly remove solid, well documented studies because the numbers they show are uncomfortably high. You have done this by pushing the exclusive use of the narrowest definition possible, which you claim to be universal but have yet to provide any source to prove it.--DeanKosta (talk) 21:47, 24 July 2013 (UTC)


 * Well, let's take a step back for a moment: we all have our biases, but that doesn't mean we need to assume bad faith on each other's parts. As I see it, the important thing here is to clarify how the term has been used by various studies or commentators. That's a reasonably objective thing that we should all be able to agree on. If indeed, Greer was the only one to apply the term to misidentifications (although he apparently didn't think he was, because he was critiquing the studies performed by other people), then we can note that. If not, then we can alter the definition section to provide a summary of how various people have applied the term. JudahH (talk) 05:04, 25 July 2013 (UTC)


 * About the 23 or 26 percent thing, by the way, it's important to note that these percentages refer to the cases revisited in light of DNA evidence not all accusations in general. I agree that these studies should appear somewhere in the article. JudahH (talk) 05:16, 25 July 2013 (UTC)


 * DeanKosta - I've never worked on False accusation or False accusation of child sexual abuse, but it is possible that for child abuse, different classifications are used because hey, children are not adults. We need to go with the definition that's used by reliable sources, not the definition that you think is logical based on definitions of other things. (FYI, all of the sources in the section of False accusation that you linked are about child abuse.) I've already mentioned that I need to look into whether or not deliberate misidentification (referred to as "perpetrator substitution" for child abuse, but not for rape) is considered a false accusation in the literature, but that appears to be the only substantial difference; your laundry list of other things that count, eg. suggestive questioning or otherwise poor interviewing, are means by which false allegations come about, since children are suggestible, not other categories we've left out.
 * Judah H, re 23/26% - yeah, I know that that's as a percentage of exonerations, not of rape charges - however, the analysis combines numbers for perjury by other people, deliberate misidentification, and no crime. As I said, I'm not sure about deliberate misidentification counting or not and have to look into sources more, but perjury by others doesn't fall under this topic. –Roscelese (talk &sdot; contribs) 15:23, 25 July 2013 (UTC)

Differences among the various studies and sources cited in the article
As I mentioned above, I think it would be worthwhile for us to compile information about the various studies cited here: their definitions, assumptions and methodologies. As a start, I've copied the table given in the article to this Talk Page and expanded it so that we can add the given information where available. That way we'll have a clearer picture of what all these sources are really saying:

Obviously, there's more work to be done before this might be useful, but I thought it would help to set up a framework for organizing such information. Feel free to edit the above table if you have better ideas on how best to organize it. JudahH (talk) 18:07, 24 July 2013 (UTC)

sections
I hope I don't have to copy/paste examples (i will though if i need to) of the the organization of "Estimates of the prevalence of false accusations" with most of the studies discussed have the criticism along with the main description while the criticism for the "Kanin" section is separate with it's own subheadNailo1234 (talk) 18:59, 10 July 2013 (UTC)

On the scientific validity of Lisak's study and methods
David Lisak's study is presented in such a way that it is being used to suggest that allegations that aren't shown to be false with evidence should be assumed to be true. In science, a statement usually is considered false without evidence supporting its truth. No evidence should be required to suggest that a claim is false when there is insufficient evidence establishing a claim's truth. This isn't intended to be a personal attack on Lisak or a discrediting of rape allegations, this is just a criticism of Lisak's methods. Astrohoundy (talk • contribs) 21:53, 14 July 2013 (UTC)
 * If "Lisak's study is flawed in this way" isn't just your personal opinion, you should be able to find the same opinion in a reliable source. It is actually quite simple. –Roscelese (talk &sdot; contribs) 23:29, 14 July 2013 (UTC)

I found a source that challenges the validity of IACP guidelines a while ago and I posted it, but mention of it still follows mention of DiCanio's listing. I think it should be placed ahead of DiCanio's study. Astrohoundy (talk) 16:54, 14 August 2013 (UTC)
 * Is this in the Greer source, do you mean? –Roscelese (talk &sdot; contribs) 18:09, 14 August 2013 (UTC)

Weasel words
I tagged the following paragraph as containing weasel words: "Rates of false accusation may have been misrepresented as larger than they are. A rape case may be closed as "unfounded" without a conclusion, but there are many reasons other than falsity for which a report that is not false may be classified as "unfounded."'

This is an awful introductory paragraph for a number of reasons: 1. It attacks a position without identifying that position. 2. "May have" are weasel words. 3. Its second sentence makes a point that is irrelevant, since, again, the alleged misrepresentations have not been identified or discussed. Until you can point to someone calculating false accusation rates by "unfounded" classifications, the sentence has no place here. 4. It's true that rates of false accusation may have been misrepresented as larger than they are. It's also true that they may have been misrepresented as lower than they are. In fact, considering the range of estimates, BOTH are almost certainly true. Discussing one without the other, implies that these rates tend to be overrepresented rather than underrepresented. Considering that all numbers in this area are estimates, it's very hard to see how such a claim can be proven. 5. The paragraph is poorly sourced: the first source links to an entire book, and the second source redirects to a blank search page. It's impossible to see what the sources for this paragraph are supposed to be. 6. In any case, the paragraph is gratuitous. Simply quoting the various studies and estimates on the matter is sufficient, without a biased editorial comment to introduce them.

At first I simply tagged the more problematic words of this paragraph ("may have", "misrepresented" by whom?) but these tags were deleted by editor Roscolese, who, judging from his numerous edits to this article, has an agenda to push. At this point, I am removing the paragraph for the reasons cited above. JudahH (talk) 23:45, 18 July 2013 (UTC)


 * Various responses: The sentence perfectly reflects what is stated in the source. We can't make our own decisions about what it really means or about whether we think the evidence supports it. The paragraph is appropriately sourced and your inability to access the sources doesn't affect their reliability; I'm looking at page 185 of the book and a cached version of the article. It's not intended as an introduction to the studies, but rather as a reliably sourced fact about the prevalence of false rape accusation. –Roscelese (talk &sdot; contribs) 02:06, 19 July 2013 (UTC)
 * OK, first of all, in terms of sources: a reader's ability to access the sources may not reflect their reliability, but they do reflect their verifiability, to the reader, at any rate. Thanks for clarifying what page the material was from (maybe that information should be in the footnote). As for the second source, apparently you need to have a subscription to something for that to work; it would be better if a free version were available, but if one isn't, would you quote the relevant part, please?
 * Second, sourced or not, it's an inappropriate sentence to pluck out of a book and use as the introduction to a sentence. We do have to make our own decisions about what information to present and how to present it, and I believe this is a poor choice for a number of reasons. 1) It's skewed to one side. I have little doubt that I could find a source to state that "the frequency of false rape accusations may often be understated". If I found such a source, would it be appropriate for me to use that as the introductory sentence instead? Introductions should be balanced, and this one is not. 2) It's vague. It doesn't help to have a reliable source if that source is making only the vaguest of statements. I've read the original source now, and I'll copy the paragraph here for reference:
 * Because of this method of "closing" an investigation, rates of false allegation may have been inflated and misrepresented. For example, it may be reported that false allegations of rape occur at the rate of 30%, when what is really meant is that 30% of cases have been closed as "unfounded
 * This is not making a definite statement about anything. It's saying "this may plausibly have happened in some places at some times". It's simple speculation. I'm not saying that this doesn't have a place in the article at all. It could be included as part of a discussion of what goes into counting false allegations, and the classification of "unfounded" being one potential pitfall. However, as an introduction to the whole topic, I believe that it is one-sided and misleading.
 * As a sign of good faith, I won't take the paragraph out again before giving you a chance to respond, but I honestly believe that this paragraph should not be the introduction. JudahH (talk) 05:39, 19 July 2013 (UTC)
 * Like I said - We can't do our own background research to find out which studies (or even which popular ideas) the source may be talking about. It's not an introduction, but simply a fact cited to multiple sources that doesn't belong in any subsection since those are all about particular studies; your insistence on misunderstanding as an excuse to remove sourced material is childish. –Roscelese (talk &sdot; contribs) 15:39, 20 July 2013 (UTC)
 * Interpreting a source based on its own words is hardly "original research". The source, which I quoted above, does not refer to any studies in particular, but to a hypothetical example. Meanwhile, you may claim that it's not an introduction, but its placement at the beginning of the section and the general nature of its statements certainly makes it read that way.
 * Judging from this talk page, a consensus seems to be emerging against you: most editors who have weighed in agree that this paragraph's tone and placement are inappropriate, yet you keep on reverting back to it, on your own. However, on the principle of compromise, I've refrained from deleting the paragraph again. Instead, I attempted to balance it by adding a consideration on the other side, with source. JudahH (talk) 20:35, 22 July 2013 (UTC)
 * Your "consideration" is unfortunately not reflected in the source. –Roscelese (talk &sdot; contribs) 23:10, 22 July 2013 (UTC)
 * It certainly is. Just as your source points out that a conclusion of "unfounded" does not necessarily imply a false accusation, mine demonstrates that a conviction does not necessarily imply that an accusation was correct. JudahH (talk) 01:21, 23 July 2013 (UTC) However, I've added another source that directly addresses the issue of false allegations leading to convictions that are later overturned. The study found that 23% of such convictions involved perjury or false accusation. JudahH (talk) 01:58, 23 July 2013 (UTC)

JudahH, I agree wholeheartedly that it should either be taken out of the article altogether, or be included with other studies that are currently being removed despite having the same, or even more stringent, standards of evidence and encyclopedic content. Astrohoundy (talk) —Preceding undated comment added 12:14, 20 July 2013 (UTC)
 * Astrohoundy, you've already stated that you're coming at this from the scientifically unsound position of "all accusers are perjuring themselves," so I'm not inclined to believe this professed interest in including well-conducted studies. –Roscelese (talk &sdot; contribs) 15:36, 20 July 2013 (UTC)

I've never said anything of the sort. I've said that, in science, a positive statement is considered to be false until evidence suggests that it is true. This is the Closed World Assumption. Astrohoundy (talk) 15:43, 21 July 2013 (UTC)Astrohoundy


 * What about rape is scientific? Rape is a social problem, not a scientific problem. Binksternet (talk) 16:35, 21 July 2013 (UTC)

Binksternet wrote "What about rape is scientific? Rape is a social problem, not a scientific problem. " Any process that involves determining truth or falsehood is inherently an attempt at science. This includes the analysis of all social evils. Perhaps I was incorrect to state that any statement is considered false until evidence suggests its truth, as the Closed World Assumption isn't always used(though it is used the majority of the time), but I was abundantly correct to state that it is patently unscientific to assume that all false statements have evidence supporting their falseness, especially when there is no, or insignificant, evidence supporting statements' truthfulness. It's like this:

Consider 100 accusations made. Of these 100 accusations, 30 have strong evidence such as blood, positive DNA, and poisons like ketamine supporting their truthfulness. Say 6 have strong evidence supporting their falseness such as alibis, negative DNA testing, previous statements from the accuser indicating a likely plot to make a false accusation, and video footage. What about the other 64 accusations? There isn't much evidence that they're false, but there isn't much, or any, evidence of truth either in these cases. In this case, it would be incorrect to assert that the false accusation rate is only 6 percent, as this implies that 94% of the accusations are correct. However, there's no evidence to support the 94% accuracy rate. There's only evidence to support a 30% accuracy rate. This is the heart of the matter. The proven, or demonstrably, false accusation rate ranges from 2%-8%, which is not necessarily the actually false accusation rate.

Given all of the awards and accolades prosecutors receive for pursuing spurious rape trials often, considering the contracts Wendy Murphy got after defending Mike Nifong, considering statements often made by former prosecutors like Craig Silverman and Linda Fairstein, it's reasonable to consider this true-until-proven-false mentality to be ridiculous.Astrohoundy (talk) 17:39, 21 July 2013 (UTC)
 * I've already explained this to you and it's also in the article text. Making false accusation is an action that someone commits, not an abstract concept floating around in the ether; what you're saying is that, without any actual evidence, we must nonetheless believe that the accusers are perjuring themselves. There's a reason "unfounded" is a category at all, and it's because sometimes there's not enough evidence one way or the other. –Roscelese (talk &sdot; contribs) 20:23, 21 July 2013 (UTC)

Just a passerby arriving here via the OR noticeboard. There are some things that are very unclear in the article that may be why this discussion is so difficult to have (and I apologize if I'm way off base -- I haven't been to this page before). Can I suggest a section right at the top called something like "definitions" or something to the effect of "research difficulty"? The current list of studies is kind of unwieldy, with several sections offering similar evaluations and criticisms but without clear connections between kinds of criteria, data, and measurements used. A section explaining the different kinds of criteria, separating legal from common use terms, and explaining the kinds of entities doing the studies would be very helpful (e.g. the police surveying their own data, lawyer reviewing court cases, sociologist interviewing defendants, etc.). It seems the authors of these studies have a lot to say about methods they and others use which would make for a useful lead-in to the statistics themselves. The "Other Studies" section does not contain studies like the others and its contents could simply be folded into this introductory section. At the very least, it reduces the most controversial aspects of this article to one section and allows the various findings to stand on their own without selective responses and criticisms (except perhaps where particularly notable).

The reason I say the lack of clarity may be part of this discussion is because it seems like scientific and legal terms might be getting conflated at times and contrasted at others. Our legal system can be understood as scientific, but "Innocent until proven guilty in a court of law" is not the same as saying the events themselves did not actually happen in the world until a judge or jury declares it so. It's a question of criteria/data. If the measure is based on legal filings, you have a statistic about what happened in court, which is open to criticism for not telling the whole story. If the measure is based on interviews with accusers/accused or other means, you have a different kind of figure subject to its own criticisms. All that could be explained up front and the reader primed to better understand all of this great information you've compiled. --Rhododendrites (talk) 02:47, 22 July 2013 (UTC)
 * The idea of a section on research methods is interesting and potentially useful, but your conflation of definition with research methods is less appealing. The definition is not at issue; regardless of what certain unconstructive editors would like you to believe, it is an accusation by a supposed victim of a rape that did not happen, and we won't waste space on whatever poorly sourced "all accusations are false" or "incorrect DNA identification proves that women are lying bitches" nonsense is out there. (I'm not accusing you of this, I just happen to have been keeping an eye on this article for a while.) I'm somewhat confused about your last paragraph. Can you clarify what sort of material you are suggesting adding? –Roscelese (talk &sdot; contribs) 03:58, 22 July 2013 (UTC)
 * Hmm. I'm not sure why the conflation of definition and research methods is mine. While the definition may be simple and clear to you and is clearly stated in the first sentence, the rest of the article complicates it in a steadily increasing litany of definition, legalese and legal definition, data sources, methodological language, and constraints and criticisms of all of them. I'm not supporting basing any of this article on the above arguments which seem to boil down to tenuous extensions of "innocent until proven guilty" because doing so turns a discussion about what things happen in the world into one about jurisprudence. My point was only that having a section up top explaining the studies and the terms they use would make it not just a clearer article but also easier to prevent (or wrangle) some of the talk page rhetoric.--Rhododendrites (talk) 05:33, 22 July 2013 (UTC)
 * Can you explain what it is that you're confused about? Nothing seems unclear to me about the various studies - in a given set of accusations, some are provably true, a few are provably false and thus the subject of this article, some others can be proven neither false nor true (= not enough evidence to convict the accused, but also not enough evidence to convict the accuser). A methodology section, as you suggest, might be useful for collating information about how such studies/statistics are conducted, but the actual definition of a false rape accusation is not at issue - an accusation by the supposed victim of a crime that didn't happen is the definition used in all the relevant literature. –Roscelese (talk &sdot; contribs) 05:56, 22 July 2013 (UTC)

"Can you explain what it is that you're confused about? Nothing seems unclear to me about the various studies - in a given set of accusations, some are provably true, a few are provably false and thus the subject of this article, some others can be proven neither false nor true (= not enough evidence to convict the accused, but also not enough evidence to convict the accuser)." It is this third category that is the point of much contention and conflict. If the article emphasizes that the studies in question only dealt with the demonstrable false accusation rate, but presents the results in such a way that it suggests that over 90% of rape accusations are true. It appears to me that this is based on a desire to see simply see the accused convicted in courts on little more than accusation alone. — Preceding unsigned comment added by Astrohoundy (talk • contribs) 17:48, 26 July 2013 (UTC)
 * You're asking us to engage in original analysis, and that's a request we simply cannot accommodate. If it's a question of providing the percentage of "unfounded" reports in addition to the percentage of false ones when the sources provide the split, that's one thing, but your repeated insistence that we insert your own personal views on the sources is not going to get you anywhere. –Roscelese (talk &sdot; contribs) 17:53, 26 July 2013 (UTC)

No, I'm not. That isn't original analysis. Simply reading the study to determine the standards of evidence used to estimate the percent of allegations that are true is not original analysis. The article needs to clarify that many of these studies assume allegations to be true unless evidence suggests them to be false, which is not a scientifically sound position.--Astrohoundy (talk) 19:54, 2 August 2013 (UTC)
 * "reading the study to determine the standards of evidence used to estimate the percent of allegations that are true" is precisely original analysis. Anything you read into it that isn't stated in the source is original analysis. I've already addressed, repeatedly, your false claim that reports which are not proven false are assumed true and subsequent fallacious contention that there is something scientifically unsound about not making claims about perjury people have committed without any evidence. –Roscelese (talk &sdot; contribs) 20:48, 2 August 2013 (UTC)

FBI Results 2006
I quote: "Every year since 1989, in about 25 percent of the sexual assault cases referred to the FBI where results could be obtained (primarily by State and local law enforcement), the primary suspect has been excluded by forensic DNA testing. Specifically, FBI officials report that out of roughly 10,000 sexual assault cases since 1989, about 2,000 tests have been inconclusive (usually insufficient high molecular weight DNA to do testing), about 2,000 tests have excluded the primary suspect, and about 6,000 have "matched" or included the primary suspect.1 The fact that these percentages have remained constant for 7 years, and that the National Institute of Justice's informal survey of private laboratories reveals a strikingly similar 26-percent exclusion rate, strongly suggests that postarrest and postconviction DNA exonerations are tied to some strong, underlying systemic problems that generate erroneous accusations and convictions."

From this we see consistently a 20% minimum false accusation rate. This does not mean false accusation by person raped necessarily, but false accusation rate by the justice system certainly. Reference may be found at the National Criminal Justice Reference Service https://www.ncjrs.gov/txtfiles/dnaevid.txt SunSw0rd (talk) 00:04, 26 August 2013 (UTC)
 * Please read the sources and the previous discussions instead of making us go through all this again. Incorrect identification of a suspect is not part of this topic, even if you coin a phrase "false accusation by the justice system." –Roscelese (talk &sdot; contribs) 02:50, 26 August 2013 (UTC)


 * Agreed Thereandnot (talk) 12:59, 26 August 2013 (UTC)


 * Would you prefer "erroneous accusation"? The error rate is valid. SunSw0rd (talk) 21:22, 17 September 2013 (UTC)

Edit request on 24 September 2013
Dear Author(s) of this page on false rapes. I'd very much appreciate if you could add to the "Further Reading" a bullet with my article in the same edition as Lisak's et al. this is the cite adn I'm happy to send you a pdf of the article if you wish:

I am the president-elect of the American Society of Criminology, and become the president the end of November 2013. Here is my website: http://www.colorado.edu/ibs/pb/belknap/

Thank you! Joanne

198.11.24.213 (talk) 14:34, 24 September 2013 (UTC)


 * ✅. Thank you for your note! And congratulations on the upcoming presidency. Binksternet (talk) 19:07, 24 September 2013 (UTC)

Anglophone bias
The article is highly distorted towards the Anglophone point-of-view, as only the US, Australia and the UK are included. I'll start by adding information about false rape accusations according to the Finnish police (from 2010). Of course the only sources are not women's studies scholars but national police as well. It would be great to have information from more countries included! --Pudeo' 19:02, 25 February 2014 (UTC)

Police in Finland
User:Pudeo added a section on police estimates of the prevalence of false sexual abuse allegations in Finland. The source for the section is an article in the Turun Sanomat. However, the article does not mention an actual study or investigation concerning false rape allegations. It says that "Officials estimate that more than one in five allegations of rape to the police turn out to be false" but it doesn't say which officials and what data these estimate are based on (e.g., what period of time, Southwest Finland or all regions of the country?) or what criteria were used for distinguishing a false allegation from an unfounded one or one lacking evidence etc. No official report, or study, or investigation was presented. The claim that the prevalence of false rape accusations has risen with the prevalence of alcohol consumption by women is based on a statement from one police officer. The opinion of one police officer hardly qualifies to be included as an official estimate in this article. I suggest that the section be deleted until the actual police report (if it exists) is presented here. --Sonicyouth86 (talk) 12:36, 18 April 2014 (UTC)

Contradictions
Did anyone bother reading Section II, Crime Index Offenses Reported? I did, and the report, without question, does not refer to false accusations. The report indicates unfounded rape accusations, which are quite different. Which by the way, if anyone had read the report, they'd know that the 2% number is the median average for all index crimes. Saying the rate is between 2% and 8% is not only false, the study being referenced actually directly states thus. In reality, the FBI's DNA exclusion rate, as reported by the DOJ, between 1989 and 1996 was about 24%. Why was this information not included? Why has the sentence blatantly contradicting the report (the 2% number) not been changed? Why has the FBI's own DNA exclusion rate not been included under a section entitled "FBI statistics"? We have the FBI's statistics, they were compiled, and published, by the DOJ. Why is that information not included, while patently false (the 2% number which if obvious if anyone read the report) and misleading (the 8% unfounded rate) were included?

To quote the report: "Eight percent of forcible rape complaints in 1996 were “unfounded,” while the average for all Index crimes was 2 percent." Someone just lied when they made that statement, obviously. — Preceding unsigned comment added by 108.25.51.70 (talk) 23:22, 27 April 2014 (UTC)

Amazing, the entry for rape in Wikipedia actually addresses the issue well, and is well referenced. It of course directly contradicts this article though. http://en.wikipedia.org/wiki/Rape#cite_note-109 — Preceding unsigned comment added by 108.25.51.70 (talk) 23:46, 27 April 2014 (UTC)

P.S. - Almost forgot. If anyone had actually researched the issue in detail, utilizing the FBI database, they'd know that the FBI report, False Allegations of Adult Crimes, by James McNamara, M.S., and Jennifer Lawrence, M.A., in fact uses Kanin's research, which was reviewed and confirmed by Carney, along with McNamara and Lawrence, according to FBI methodological practices. — Preceding unsigned comment added by 108.25.51.70 (talk) 23:29, 27 April 2014 (UTC)

This article contains two contradictory claims:

"The typical person making a false accusation is a white woman between 21 and 30 years of age." (section Profile)

vs.

"About half of the cases involved people aged 21 years old and under" (section Crown Prosecution Service report (2011–2012))

The part with "white" seems also quite doubtful - see Duke lacrosse case - false accusations made by Crystal Mangum.

A Reader (talk) 12:18, 1 November 2013 (UTC)


 * This is presumably the result of the two statistics coming from different locations, but yeah, that ought to be clearer. Mangum doesn't disprove anything, as she is, obviously, only one person. –Roscelese (talk &sdot; contribs) 14:31, 1 November 2013 (UTC)

The statement "According to surveys of American law enforcement officers, the typical person making a false accusation is a white woman between 21 and 30 years of age." has been edited from the cited source. That is based on a single survey of 20 law enforcement officers with a combined 560 years of experience. What they stated is that about 65% of those making false accusations were between 21 and 30 years of age (and 100% of those making false accusations were white females). However the statistics also noted the following: of 33,360 violent crimes of ALL types 492 dealt with false accusations of sexual assault. Assuming that rape is about 7% of all violent crime (current FBI statistics), applying that 7% to the 33,360 then 2,335 of those crimes would probably have been rapes. Adding 492 to 2,335 (because assuming that those were subtracted from the total number of rapes) and then calculating the percentage of false accusations against the total (492/(2335+492)) results in a 17% false accusation rate. Of course the problem with this calculation is that (a) it uses a survey of 20 officers, then (b) uses the FBI statistics to calculate the rate of rape against all violent crime, and then (c) again uses the 20 officer survey for the number of false rape accusations. So my point? If you don't think 17% is a credible number for false accusations, then I suspect you won't buy that only white women make false rape accusations and that 65% of them are between 21 and 30. However the citation to the source is valid follow the link and then click through the pages to page 190. That doesn't mean the survey referenced at that point in the book is valid. SunSw0rd (talk) 20:04, 20 December 2013 (UTC)
 * Puzzling through your somewhat incoherent comment, it seems like you're trying to do your own original analysis, which is not permitted. –Roscelese (talk &sdot; contribs) 22:44, 20 December 2013 (UTC)
 * This is a talk page (commentary) -- not the article -- nothing wrong with analysis in the talk section. I am pointing out that (a) the bottom section of the article titled "Profile" is not quoting from the cited source and is misleading, and that (b) the source of the first sentence in "Profile" is from a limited survey of 20 police officers, and that (c) applying the above statistical analysis it may be suspect. Actually since the source states that 100% of the false accusers are white women it should be obvious that it is suspect. Clear now? SunSw0rd (talk) 00:56, 21 December 2013 (UTC)
 * What changes are you suggesting? The talk page is not a forum. –Roscelese (talk &sdot; contribs) 17:27, 21 December 2013 (UTC)
 * Correction supplied -- changed the inaccurate 1st sentence to now being accurate by clearly referencing the fact that it is not "surveys" plural but a single 2004 survey of only 20 officers and (b) replacing the inaccurate generalized sentence by providing a direct quote from the source. SunSw0rd (talk) 03:03, 23 December 2013 (UTC)

Semi-protected edit request on 21 May 2014
"According to a single survey of 20 American law enforcement officers done in 2004, the typical person making a false accusation was ..."

This is not a survey of perpetrator data. It is a survey of perceptions of law enforcement officers. The source material says this: "To the best of the authors' knowledge, there are no statistics maintained on false allegations." And of their survey: "It is to be noted that this survey was not intended to provide detailed predictive analysis, but to report investigative perceptions." Wiki's reporting needs to stress this. It's an important distinction because perceptions aren't necessarily realistic.

I'd also question whether this deserves its own subsection or mention. The article is about false accusations, not about public perceptions of false accusations. Making this the only reference to (perceptions of) perpetrator data might be giving this single survey (of only n=20) too much weight. Maybe add more data about perceptions or actual demographic data?

Proposed rewrite:

It is unclear if there are any statistics maintained on demographics of false allegations, but a survey of 20 American law enforcement officers done in 2004 reported their investigative perceptions:


 * The officers reported that, in their experience, the person typically making the false allegations was female (100%), Caucasian (100%); 15-20 years of age (10%), 31-45 years of age (25%), or 21-30 years of age (65%).

A false accusation may be perpetrated out of a desire for attention or sympathy, anger or revenge, or to cover up behavior deemed "inappropriate" by their condemning surrounding culture.

Emarkcd (talk) 08:03, 21 May 2014 (UTC)


 * Red information icon with gradient background.svg Not done: It isn't obviously WP:UNDUE, but you could try to gain consensus that it is. I'm concerned that your suggested change is moving away from a neutral statement of the facts. Too many weasel words give the impression that we disagree with the results of the, admittedly small, survey. (It is already a little non-neutral that we currently say a 'single' survey, when we are clearly talking about one survey of 20 police officers in 2004.) A better approach would be to find a larger study of the demographics of people making false allegations and add that content to this section. Regards, Older and ... well older (talk) 19:39, 21 May 2014 (UTC)

Something missing not sure how to write it
Reading this article you come to the conclusion that 2-8% of all allegations are false and therefore 92-98% of allegations are true. We could do with finding some sources that are explicit on the middle ground, the undetermined majority of cases that make the numbers limited to the point of meaninglessness.

In studies various numbers from 1.6% to 20% (that I have seen) of cases have met the standards of false. [In many cases this includes only knowingly false and in others in includes things the officers strongly suspect are false] Statistics also show that of reported rapes in the region of 6.7% result in prosecution (Universiyt of Kentucky 2011). These are the two comparable figures because the remaining 88.3% of reported rapes are indeterminately true or false. Not to say all 5% of reports determined false are false or all convictions are correct but they at least have been determined to some scrutible standard.

Interesting point (Entirely OR and not suggested for the article), if the ratio of "determined false" and "Determined true" persists throughout the undetermined then it comes to a figure close to 41% which was the figure Kanin came to. — Preceding unsigned comment added by 213.106.233.97 (talk) 13:42, 12 June 2014 (UTC)

Police in Finland
User:Pudeo added a section on police estimates of the prevalence of false sexual abuse allegations in Finland. The source for the section is an article in the Turun Sanomat. However, the article does not mention an actual study or investigation concerning false rape allegations. It says that "Officials estimate that more than one in five allegations of rape to the police turn out to be false" but it doesn't say which officials and what data these estimate are based on (e.g., what period of time, Southwest Finland or all regions of the country?) or what criteria were used for distinguishing a false allegation from an unfounded one or one lacking evidence etc. No official report, or study, or investigation was presented. The claim that the prevalence of false rape accusations has risen with the prevalence of alcohol consumption by women is based on a statement from one police officer. The opinion of one police officer hardly qualifies to be included as an official estimate in this article. I suggest that the section be deleted until the actual police report (if it exists) is presented here. --Sonicyouth86 (talk) 14:18, 30 June 2014 (UTC)


 * You've posted this thread before without objection and seem to have examined the sources (I have not). Is there a reason you haven't just made the edit? –Roscelese (talk &sdot; contribs) 21:28, 30 June 2014 (UTC)
 * I wanted to give Pudeo time and an opportunity to provide the actual police report or statistics rather than a newspaper article but I can see now that s/he doesn't know the source of the supposedly official stats. So I'll remove the section for now. --Sonicyouth86 (talk) 13:05, 1 July 2014 (UTC)


 * Sorry, I didn't see your earlier thread because apparently it was instantly archived. I think it's pretty clear from the article that it deals with whole Finland and that the 1 out of 5 number comes from official police statistics. The comparision with risen alcohol consumption and most false accusation "suspects" being immigrants are statements by police officers, but it's not like they would just tell their personal opinions to a large newspaper in the name of their department.


 * But yes, it's a police statistics not a study. The newspaper itself is a reliable one and I think they made that article because there was a well-publicized false accusation case in the regional area back in 2010. However, that Internet version is not as extensive as the one publicized in the paper version as that webpage says "Read more from wednesday's paper". Perhaps the paper version would have more accurate information. But understandably verification from non-English sources is problematic although there is no rule of preferring English-language sources in Wikipedia.


 * As for not it being a study, I believe more diverse sources are good for the article. Atleast if one was to advance the position that only studies by the feminist school of criminology should be included, it could result in NPOV problems. --Pudeo' 22:21, 30 June 2014 (UTC)


 * The Turun Sanomat is a regional newspaper and it's unclear if the estimates are from the regional police or national police estimates. If the 1 out of 5 number comes from official police statistics, it should be possible to provide the police statistics rather than a newspaper article that doesn't include fundamental information like the time period for these estimates (did they look at reports between 1999 and 2002 or 2005 and 2006?), the number of reports (is one-fifth out of 13000 reports false or one-fifth out of 130 reports?), and the criteria used for deeming rape allegations "false". As for your contention that we should include questionable content for the sake of diversity and in order to counterbalance "studies by the feminist school of criminology", let me ask you something: What makes you think that the Crown Prosecution Service, Australian police, the FBI and scholars like Rumney, Kanin and Lisak are from "the feminist school of criminology"? --Sonicyouth86 (talk) 13:05, 1 July 2014 (UTC)
 * I don't think it makes any difference if the newspaper is "regional", they run as many national stories as well, being a major paper. The New York Times is a local metropolitan newspaper as well, that doesn't mean their stories would only be about the New York metropolitan area. It is implicit their estimation is from rape reports from 2009-2010 (why would they estimate history and run it as news?) The news story is listed in their "National news (KOTIMAA)" and refers to "the police" (there is only one national police in Finland with local branches) without even mentioning the local department so they do refer to national level. What we have here is a reliable newspaper running an article with consulting the police and their statistics. This article reads "The Finnish police estimates.." and "The Helsinki Police Department states...". Those are undeniable facts. If one thinks those statistics are flawed, I guess that's up to the reader, but we're just telling what has been estimated. If we were to say "1 out of 5 rapes in Finland are false accusations" it would be a weasel words, but we're telling who estimates and what. --Pudeo' 13:46, 1 July 2014 (UTC)
 * You may see it as "implicit", but the fact remains that the article doesn't state the time period that the estimates are based on, the "sample size", and the criteria used. It's a case of WP:OR. You claim that "the Helsinki Police Department states" when in fact it's one police officer who states that and it is absolutely unclear what his opinion is based on, a systematic investigation or personal experience. You also claim that the "official statistics" are from 2010 although the newspaper article doesn't support that. As for the 1 out of 5 number, what is that number based on? On 5 reports or on 50? In this context, the article in the Turun Sanomat is clearly an unreliable source. You have stated your reasons for including this article (something about combating "the feminist school of criminology"), but they are insufficient to include the personal opinions of individual Finnish police officers and dubious news reporting in the wake of "a well-publicized false accusation case in the regional area back in 2010". --Sonicyouth86 (talk) 14:03, 1 July 2014 (UTC)
 * My reason for including this information in the article isn't "combating" the feminist school of criminology, I just said that to acknowledge that this topic is controversial and, in a somewhat caricatured way, an MRA would make a big deal about false rape accusations and a feminist would play it down (one of the usual disputes related to these topics in WP). My actual reason for including this was to address Anglophone bias, as I stated. Currently the article only has information from the US, UK and Australia. And indeed, apprently it's hard to find accepted non-English sources. I do agree with you that it's a more light-weight source because it's an estimation by the Police - but is that enough to remove it completely from the article? --Pudeo' 14:21, 1 July 2014 (UTC)
 * Pudeo, I don't think linguistic/geographical diversity is a reason to suspend our interest in RS/verifiability. As I said, I haven't checked out the source, but I find Sonicyouth's points very convincing. Having a tiny sample size or using data from a single year is very sketchy. –Roscelese (talk &sdot; contribs) 15:28, 1 July 2014 (UTC)

I removed the section from 1 Estimates of prevalence but moved the alcohol consumption link and immigrant background thing to the section 2 Profile. The prevalance estimation (1 out of 5) is removed from the article, because number seems to be very varying from other news I searched from. I hope you are fine with the solution. Ping --Pudeo' 03:51, 5 July 2014 (UTC)

100% Caucasian
The article states "the typical person making a false accusation was "female (100%), Caucasian (100%)". How is this possible? No women of color make false rape claims? Rizzerd (talk) 10:12, 30 June 2014 (UTC)


 * It was probably a very small study. Some of these statistics shouldn't really be taken on face value, particularly ones like Kanin's with very small sample states and in areas where police methods differed from the norm --80.193.191.143 (talk) 15:12, 24 July 2014 (UTC)

The headline should be 1.5 - 41%, not 1.5% to 8%
From the article itself, one study showed a false reporting rate of 41%.

There's also NPOV here -- Kanin's criticism of that study is included, while Kanin's (rather obviously weak, since it only looked at a college campus) work is elevated, and no criticism of Kanin is included.

I suggest putting all the studies in chronological order.

I'll check back in a few days; no serious objection, I'll make the change. — Preceding unsigned comment added by Djcheburashka (talk • contribs) 17:11, 4 November 2014 (UTC)


 * The study that supposedly found 41% has been roundly discredited. –Roscelese (talk &sdot; contribs) 18:48, 4 November 2014 (UTC)
 * Consider this two "serious objections".  Eve rgr een Fir  (talk) Please &#123;&#123;re&#125;&#125; 20:06, 4 November 2014 (UTC)


 * I don't think that's correct. Another study and some (rather dubious) academics disagree with the 41% study.  That does not mean it is "roundly discredited," it just means some people disagree.  Wikipedia should not take sides in this.


 * The objection to the 41% study is that its a study of what proportion women who approached police were then determined (by the police) to have actually been raped. That is, of course, a problematic methodology, because the police can scare accusers off.  No-one disputes that.


 * However, the FBI study, which is quoted as saying the rate of false accusation is 8%, is actually addressed to a different issue. It considers allegations which proceeded to the investigation stage and which were then determined to be "unfounded" following investigation.  That's apples-to-oranges with the 41% study, because it disregards accusations that don't proceed to full investigation.  Surely not all persons who approach police and then don't continue were actually raped but dissuaded by police?  The FBI number also doesn't take into account persons found not guilty, dropped-charges, and cases where the person pleaded down to a non-rape charge. So its being mis-cited here.  (See page 24 of the FBI report.)


 * In fact, the FBI uses the same methodology as the 41% study, except for exclusion of dropped-early accusations. So it has all the problems of the 41% study, plus additional problems.


 * The Lisak study -- which is the one that supposedly discredits the 41% study -- uses the same methodology as the 8% number, so it has all of the same problems of both that and the 41% study, except the Lisak study was confined to a single university and the sample size was 136 over ten years which is (again obviously) meaningless. So its even worse.


 * Can we try to resolve this in an NPOV-conscious way by not privileging one study over others, not misrepresenting the FBI number, and noting the controversy without attempting to make claims about which are right and which are wrong? Right now the page is plainly in violation.


 * Djcheburashka (talk) 00:36, 5 November 2014 (UTC)


 * The lede of the article should give an overview of how the literature presents the subject, and including, without comment, outliers known for their shoddy methodology would be misrepresenting the rest of the article. –Roscelese (talk &sdot; contribs) 15:00, 5 November 2014 (UTC)


 * I don't believe your judgement that that study is an "outlier" that is "known for . . . shoddy methodology" is accurate. Do you have any references to support that there's a consensus in the literature that the 41% study is weaker than the others?  Above, I provided some explanation and discussion of why I believe you claim is false.  Do you have any response?


 * Your response appears to reinforce the argument that the page is in its current form only because of biases.


 * So far, while you and one other person announced your view, no-one has offered any authority or reasoned defense of the the page in its current form.


 * Djcheburashka (talk) 03:04, 7 November 2014 (UTC)


 * The article contains detailed explanation from several other authors (one specifically a review of the literature) as to the problems with the 41% study. –Roscelese (talk &sdot; contribs) 15:44, 7 November 2014 (UTC)


 * No - it has a conclusory response by the author of another study, which happens to have the same problem In fact the other studies don't even address the topic of the page.  I pointed that out and you haven't responded.


 * The 8% and 2% numbers just cannot be right -- in 40% of rape cases, there's either an acquittal or abandonment of the rape charge: https://www.rainn.org/get-information/statistics/reporting-rates


 * It seems that you're just repeating the bias in the article and refusing to engage on the substantive issue.


 * Djcheburashka (talk) 00:56, 8 November 2014 (UTC)


 * I'm finding more and more as I go through this. The description of the Rumney article is directly contrary to what that article actually says about the Kanin study.


 * Djcheburashka (talk) 01:15, 8 November 2014 (UTC)


 * I've now edited this to take-out some of the most egregious misrepresentations of sources and biased language. I've attempted to do this in a neutral way.  This article can be far better if we work together to make it neutral.


 * Djcheburashka (talk) 02:17, 8 November 2014 (UTC)


 * Another example: The table.  This is lifted from a page of Rumney that identifies the proportion of allegations that, in each study, were *proven false/malicious*, according to *Rumney*.  For example, this table redefines "false" in a different way from the studies such that, for example, any case where an accuser withdrew her accusation is counted as "true."  Any case where there was insufficient evidence of a rape is counted as "true."  That's absurd!  The actual numbers reported in those studies, which are set forth elsewhere in Rumney, are very different.  (Some examples below).


 * Clark and Lewis: The table says 10.3%.  Actually they reported false-report rate of 64%.  10.3% is the number proven false/malicious after removing allegations that were recanted or found not to constitute a crime, etc.


 * Smith: The table says 3.8%.  Actually, Smith reported that 215/447 accusations were found not-good in one way or another.  101 lacked sufficient evidence, and 91 were withdrawn.  The 3.8% number refers to the number that were *not* withdrawn AND where there *was* sufficient evidence AND the police ultimately found the accusation had been made maliciously.


 * Harris & Grace: The table says 10.9%.  Actually, they reported 56%, with 10.9% proven false and malicious.




 * Looking at these studies, they are actually remarkably consistent. There isn't much deviation at all.  Somewhere between 40-55% of initial accusations do not proceed to an investigation, either because the complainant withdrew, or didn't cooperate, or reported facts that did not constitute a crime.  Between 8 and 20%, authorities decide to not pursue after investigation.  In a smaller number, 1.5-6%, the accusation is found provably false and malicious.  And none of the numbers take into account the proportion of allegations in which the authorities proceed to accuse someone of a crime, and the accusation then fails in the courts.  All such cases are being mis-counted, either in the original or misrepresented in the article, as "true" accusations of rape.


 * Djcheburashka (talk) 04:59, 8 November 2014 (UTC)


 * The literature defines a false accusation as intentional, not just a rape report that doesn't result in successful conviction. So what you're saying is that the table accurately represents the findings of the studies. –Roscelese (talk &sdot; contribs) 05:04, 8 November 2014 (UTC)


 * That's not correct. *Some* of the literature prefers that definition of "false."  Numerous sources take the opposite view -- as cited in the article and above.  (Which you continue to ignore.)


 * There are two streams in the literature, as cited in the article. One uses the definition that "false" means "incorrect," and finds numbers around 40%.  The other uses the definition "true unless proven to be an intentionally false and malicious accusation," and cites numbers of 2%-8%.


 * This is obviously an unresolved controversy. If you cite one side consistently to say that the other is "resoundingly criticized," that doesn't prove a point, it just establishes an NPOV violation.


 * And by the way, the lead author in the stream that you prefer, Lisak, also published studies in which he claimed, for example, that 16% of men self-report as rapists (by his definition), and that 90% of rapes on college campuses are committed by serial rapists. He also continues to assert that cases which this page acknowledges are demonstrably false, such as the Duke case, remain uncertain.  This is not a neutral source of such reliability that he should be privileged vis a vis other researchers.  He's a partisan whose work is accepted by one side of a divide and rejected by the other.  (And frankly seems as much of a quack as Stewart, etc.)  I'm not trying to take him out -- I'm just trying to make the page neutral on the controversy.


 * I really don't get why you have so much trouble accepting this.


 * Djcheburashka (talk) 07:25, 8 November 2014 (UTC)


 * The study authors evidently understand that the right way to do science is not to decree that any evidence which isn't A therefore is and must be B. If you think that a false accusation should mean any report that fails to secure a conviction, you should become an influential social scientist and do important studies that change the standards of the field, rather than trying to impose your personal interpretation of the studies of others through Wikipedia. –Roscelese (talk &sdot; contribs) 14:18, 8 November 2014 (UTC)


 * That's nonsense. If there are two possibilities, "true" and "false," and some claims are neither proven true nor proven false, its not science to put all the "unproven" ones into the "true" category.  No respected social scientist would do that, and you are not qualified to be saying what are and are not the standards of the field.


 * Notably, on the Lisak page, you persist in keeping misrepresentations regarding Lisak's qualifications and background. Here, you insist on keeping misstatements regarding sources.  At no point do you ever attempt to justify your claims.  This is a class POV problem.


 * In particular, your view is that when calculating the percentage of accusations that are "false," any time an accusation is recanted, it should be coded as a "true" accusation. That is crazy.  It is not the standard in the literature.  It is Mr. Lisak's standard and your own.  That's it!  Djcheburashka (talk) 19:29, 8 November 2014 (UTC)


 * These studies are measuring the number of false accusations according to the definition that the field uses, not measuring the number of ones which aren't proven true. Again, if you want to change social science standards, Wikipedia is not the venue for it. Now, do you have any constructive comments to make, or is it time to remove the POV template, given that "the page doesn't reflect my personal interpretation" is not an actionable POV issue? –Roscelese (talk &sdot; contribs) 19:53, 8 November 2014 (UTC)

That is not the "definition the field uses." More than half the sources in the article use a different definition, that "false" equals "not true."

Your refusal to make this page NPOV is only a demonstration of bias. I recognize that you will never accept reality -- that's why its being taken out of your hands.

Djcheburashka (talk) 00:01, 9 November 2014 (UTC)

NPOV tag
I don’t understand the rationale for the added POV tag. Previously, I’ve seen arguments that rape cases where no conviction is secured should be counted as “false accusations of rape” and I gather because the article does not do this, it is being argued the article is not NPOV. This does not seem reasonable to me. --BoboMeowCat (talk) 20:03, 24 November 2014 (UTC)


 * The talk section above this one clearly shows why this page is not NPOV. This has also been demonstrated via the disingenuous reasoning given for a recent reversion (editor tagged as 'editorialising' but removed the whole section, not just the part they had flagged as editorialising, in a manner that removed an important and valid balancing statement that derived directly from the article cited). This page more than qualifies for the lack of NPOV flag as it stands and is being actively policed to keep it so. Theduinoelegy (talk) 21:21, 24 November 2014 (UTC)


 * My reverting you here has nothing to do with this article supposedly not being WP:Neutral, unless, of course, it's your addition that was not WP:Neutral. But I highly doubt that you understand what WP:Neutral means; as has been stated at the WP:Neutral talk page, far too many Wikipedians don't understand the WP:Neutral policy; being neutral on Wikipedia is not what it means in common discourse. And as for this bit you re-added, I'll leave someone else to handle that. I have enough controversial topics to deal with; I certainly don't need the drama that ensues at this article. Flyer22 (talk) 21:36, 24 November 2014 (UTC)

Pretty good summary references
Slate 2009, Slate 2014. - Peregrine Fisher (talk) 02:17, 7 December 2014 (UTC)

The overview section tells a lie
The overview section says that the FBI reports that 2% of rape accusations are false. That isn't true. The FBI says 2% of rape accusations are proven false. I don't know whether the author is misquoting the FBI or wikipedia is misquoting the author, but either way the result is wrong or misleading at best. The FBI notes that higher than the 2% are likely false. It gives the unfounded number (8%) but notes some problems with the methodology. This is included later in the article. As anybody knows, the number of people convicted of committing a crime is not the same as the number of people who commit a crime. 2% should be treated as a minimum value, not as a best guess. That number (given by the FBI) is 8%. I removed the incorrect claim but was reverted. I don't have the source but I likely suspect it is being misquoted by wikipedia rather than the author misquoting the FBI. (The justice department numbers being of course, the same thing as the fbi numbers in this case) — Preceding unsigned comment added by 24.17.227.216 (talk • contribs)
 * I do have the source, and Anderson and the Justice Department are not being quoted inaccurately. The source goes on to discuss the FBI's "unfounded" number and methodology issues, just as we do. I'm not sure what you're suggesting we do; if the sources don't give us a number for "actually false" vs. "proven false", we're not going to speculate upon it. –Roscelese (talk &sdot; contribs) 23:36, 7 December 2014 (UTC)


 * Sorry, the fbi and justice department releases a report every year. The summary section should not display the opinions of one source to the exclusion of the other sources. The 2% number is prosecuted and convicted, the summary section gives the exact opposite impression. There is no reason for that source to be so priviledged in the summary section. It should be removed to the simple description with a range. — Preceding unsigned comment added by 24.17.227.216 (talk • contribs)

"Every year since 1989, in about 25 percent of the sexual assault cases referred to the FBI where results could be obtained (primarily by State and local law enforcement), the primary suspect has been excluded by forensic DNA testing. Specifically, FBI officials report that out of roughly 10,000 sexual assault cases since 1989, about 2,000 tests have been inconclusive (usually insufficient high molecular weight DNA to do testing), about 2,000 tests have excluded the primary suspect, and about 6,000 have “matched” or included the primary suspect.1 The fact that these percentages have remained constant for 7 years, and that the National Institute of Justice’s informal survey of private laboratories reveals a strikingly similar 26-percent exclusion rate, strongly suggests that postarrest and postconviction DNA exonerations are tied to some strong, underlying systemic problems that generate erroneous accusations and convictions." https://www.ncjrs.gov/txtfiles/dnaevid.txt The above poster raises a valid point. — Preceding unsigned comment added by 108.25.112.234 (talk • contribs)

Misinformation, article really needs cleaned at least at overview
I'm new to editing and I don't want to make mods angry at my lack composition, but kids don't read and are likely to throw this in each others faces at fact. First statements attributing Michelle J. Anderson for that 2% figure are not factual and are based from a report made over 25 years ago by a squad of NYPD officers not trained in statistical analysis which was initially quoted by one Susan Brownmiller in "Against Our Will" Turvey Brent is also attributed with arguing this figure in another text decrying Susan Brownmiller's work

https://books.google.com/books?id=Oge7LFaN5xYC&pg=PA398&lpg=PA398&dq=turvey+brent+cites+susan+brownmiller&source=bl&ots=h1SjOyO1KY&sig=sxMTVQ_mIZsF5uIt4V_woLfe-40&hl=en&sa=X&ei=SqSZVPyGJ8b4yQTOmoGgAw&ved=0CB4Q6AEwAA#v=onepage&q=turvey%20brent%20cites%20susan%20brownmiller&f=false

Please consider removing this 2% figure because attributing to the idea that false rape accusation don't happen presents a paradigm where there is a hole in the law that allows people to make baseless claims about the worst crime next to murder and many times the false accuser goes unpunished or with minimal impact such as wasting police time or 2 years in prison at most for filing false reports. 204.194.141.29 (talk) 17:30, 23 December 2014 (UTC) MUTT


 * It's possible we could be clearer about what we mean with Anderson's reference to "conventional scholarly wisdom", but it's not clear from what she writes that all citations of 2% trace back to this study, and the DOJ has the same figure. –Roscelese (talk &sdot; contribs) 18:23, 23 December 2014 (UTC)

Poorly Written Article
The key issue here is the tone of the article. The entire article appears to be a persuasive essay trying to discount any studies that provide a false accusation figure of higher than 2%. This is especially dubious because the majority of the studies cited on the page provide figures much larger than 2%. It's for this reason I cannot understand why:

"The "conventional scholarly wisdom," according to American law professor Michelle J. Anderson, is that two percent of rape complaints made to the police are false".

is included in the overview. First of all, it would appear from the article itself that most scholarly wisdom places the number much higher (18 of 20 studies). Secondly since when is an uncited anecdote of an anecdote a legitimate overview of an issue? So you're telling us (uncited) that Michelle Anderson says that others (uncited) conventionally (vague) state that 2% of reports are false. There is no study that even begins to indicate that! As previously mentioned, an accusation proven to be false vs. being false are two entirely different things that can't be responsibly conflated!

This paraphrasing needs to be removed as it is no where near being factual or authoritative. It is the opinion of one professor on the opinion of "scholarly wisdom" with no citation or data supporting it. — Preceding unsigned comment added by 142.129.113.158 (talk) 05:31, 12 March 2015 (UTC)