This is a straw man, although you may not realise it, and you are engaging in exactly the misuse of statistics you are accusing others of, although you may not realise it.
I think you would find if you tried to put numbers on your claims and show how they are consistent with Bayes' Theorem you would rapidly realise that you are just confused.
All you're doing is manipulating the facts to make the encounter seem innocuous, then making up numbers based on that. This is an irrelevant point.
By the way, after reading through everything, I'm betting we differ less on our estimation of the statistical frequency of assaults, than we do on 1) what frequency is necessary for Watson to be justified in her apprehension and 2) what level of interaction justifies a blanket request that people avoid approaching women in those settings.
I think you should be more careful to use analogies that are actually analogous. Making a polite sexual proposition is significantly different to "accosting" someone with a hand inside one's jacket.
You're making the same mistake that's buzzing around in this thread. I'm not generating ALL ENCOMPASSING ANALOGIES. I'm providing analogies that focus on a narrow issue. The addition of facts to a scenario greatly change the relevant statistical analysis.
That isn't a claim that the guy approaching you on the street is EXACTLY the same as the proposition on the elevator. This is just not a great way of progressing.
To remind everyone, Skeptic Ginger made a well-researched post demonstrating that if anything hotel lifts appear to be low-risk (but not zero-risk) places for rape.
You responded by saying SG was cheating by fixing on an irrelevant point, which you argued was like a tutu. Hardly anyone is robbed in a tutu, you said, but this is no comfort to someone who is robbed in a tutu.
There are multiple errors there. For one, the fact that the event took place in a hotel elevator has not only been the focus of a huge amount of discussion but it is manifestly relevant to the question of whether or not fear was justified. Trying to turn it into an irrelevant issue by comparing it to a tutu is not a rational argument.
Again, this is a childish attempt to universalize an analogy that was developed for a simple point. There's honestly no way to progress if we have to argue on this level.
The tutu example was meant to demonstrate that something that is INCREDIBLY rare statistically--someone murdered in a tutu--means very little when facts are added. It's essentially a problem of opportunity. People aren't killed in tutus very often because few people wear tutus and the people that do wear them tend to do so in a very restricted number of places.
If someone in a tutu was confronted by a person with a gun, however, what does the incredibly low rate of tutu-wearing victims tell us about that circumstance? Should the person sigh and think, "Well, I must be safe here, very few people get shot in tutus?"
The failure to understand this basic point is maddening. The tutu analogy is not meant to represent every element of the elevator situation, just that simple issue with statistics and what they tell us in real life scenarios. It's a very extreme example meant to illustrate the point. I even chose something absurd sounding to forestall the impulse to pedantically attenuate the analogy. It clearly failed.
For two, your argument would only be analogous if RW had actually been sexually assaulted, as your tutu-wearer was actually robbed. Since the whole crux of the argument was whether or not it was rational to be worried about this possibility, your analogy is not relevant.
Yes, congrats. You've noticed that analogy that wasn't meant to apply to a situation doesn't apply to that situation. Stunning work on your part.
The fact is that it would be irrational not to take the location into account when evaluating the risk of a sexual assault. A darkened car park is a much riskier place than a busy street in broad daylight right outside a police station.
Just like an elevator is a better place than a bar with lots of people in it.
If anything it's proof that some people will behave boorishly when they aren't allowed to make fallacious arguments based on irrelevant evidence, availability bias and outright mathematical errors.
No, it shows that when people have decided on a conclusion (no sexism), they will go to insanely ridiculous lengths to toss out non-conforming evidence. Sexism in academia, everyone is from academia, but the sexism stops when they go to TAM, even though the feminist skeptics have stories that match the events in academia exactly.
As I pointed out when I first raised the issue, obviously there are major differences, but you're attempt to pretend like these worlds are totally separate is just bizarre.
This is just wrong. The prior probability you will be attacked in a hotel lift is very low, as SG established. The probability that you will be attacked in the general case if someone makes a polite sexual proposition is also very low. If that's all the information we have (and it is as far as I can tell) anyone with even the most basic grip on Bayes' Theorem will see immediately that this means that the posterior probability of being attacked in a hotel lift after a polite sexual proposition is going to be at most very slightly higher than the prior probability, and is also going to be a very low probability.
Right, but we do know more. We know the woman in the elevator receives many, many threats that someone will rape her. We know the man in the elevator was present when she discussed this. We know that when they were at the bar the man ignored her request to head off to bed. We know he refused to talk to her in the bar. There was a likelihood people were drinking. You also ignored the fact that the proposition occured with no prior conversation.
You're just guessing at the stats. How many times has a strange man propositioned a woman on a hotel lift with no prior conversation? My guess is that happens in roughly 0.00001% of hotel elevator rides. We're already dealing with a subset of a subset. What percentage of those resulted in assaults? What percentage is necessary for a woman to justifiably feel uncomfortable?
Do we even need a completed assault for a woman to be justified? How many cold propositions lead to a woman being put in the position of having to deal with harrassment or just incredible annoyance? What needs to happen, how far does it need to go, before a woman is justified in her apprehension?
Assault is the worst case scenario that Watson was not in a position to defend herself against. There are many, many intermediary steps that would justify her request (and that of other women) to not be approached in that manner, but an elevator is not a place that protects a woman from assault, it's just a place where assaults rarely occur. There's a big difference between those two situations.
This is breathtaking - SG directed you specifically to
this post which makes it very clear she had already researched the subject more thoroughly than you. Cherry-picked anecdotes with no information about the base rate of hotel lift assaults are very weak evidence. As such this attempt to position yourself as someone who has done their homework more thoroughly than SG is hard to see as honest.
The methodology in that post is ridiculous. It's laden with basic errors, like the failure to look up the definition of "acquaintance" for the purpose of the statistics.
He spent three days looking for incidents on the internet. That's bad enough as a "definitive" statement on the subject without dealing with the single largest issue, which is one of reporting. We know that all instances of sexual assault are tragically under-reported. We don't know how many bald propositions resulted in annoyance and harrassment far beyond the limits of decent human interaction.
By the way, here are three assaults in elevators I found in .05sec that aren't included on that guy's list:
http://ottawa.ctv.ca/servlet/an/local/CTVNews/20110417/OTT_napanee_110417/20110417/?hub=OttawaHome
http://www2.scnow.com/news/2009/apr/16/police_follow-up_on_leads_in_attempted_sexual_assu-ar-351552/
http://www.lvrj.com/news/incident-i...d-california-tourist-in-prison-112977859.html
ETA: Here are some more:
http://www.thewesternstar.com/Justi...case-ends-with-jail-time-for-Iranian-native/1
http://psychsex.wordpress.com/2010/...with-masturbating-in-a-strip-casino-elevator/
Looks like the Ottowa case involved a serial offender:
http://www.ottawacitizen.com/news/believe+Napanee+hotel+assaults+linked/4641963/story.html
http://www.wmbfnews.com/story/13028...apping-sexual-assault-charges?redirected=true
I guess Celine Dion's husband fondled a woman in an elevator--of course, he settled and didn't admit guilt. Read into that what you will:
http://www.people.com/people/article/0,26334,623814,00.html
How many more are there? I have no idea. I certainly don't think just those that end up searchable will compose the total universe of events. We have the additional problem of knowing, for certain, that such things are under-reported. What happens when police just don't have enough evidence to go on or the case is dismissed for other reasons?
Claiming that assaults are rare, therefore she shouldn't have been afraid is based purely on ignorance concerning the nature and number of opportunities. If the conditions necessary for an assault are present, we really have no idea what the range of outcomes are and their frequency. We do know, however, that an elevator won't stop an assault.
And, by the way, I agree that all of this is rare. We just have no idea how rare and we don't know what sorts of behaviors will indicate a higher risk.
I include the above just to make it impossible for you to claim that you weren't arguing exactly this.
Yes, crazy that being in an enclosed space with someone annoying and discomforting will increase that annoyance and discomfort. I'm glad you pinned me down on that.
When in the elevator, Watson was vulnerable. There were steps she could have taken to mitigate any action taken by EG, but by that point it's already gone much too far.
Have you ever been approached by someone behaving oddly at night when you were alone? This happens with some regularity in Chicago, I don't know where you live.
The odds of me suffering any sort attack are pretty low. There's a lot of crime in Chicago, but there are millions of people and most of the issues are specific to certain areas. Any given night, my odds are low.
Nevertheless, when someone runs up to me and starts babbling, even if it's just about how the Cubs did, I become defensive. I start trying to figure out how to get away, because I honestly have no idea what's about to happen. He could just ask for a dollar or he could pull out a gun. Anything is in play.
I feel apprehension in those settings, whatever the statistics say. I would feel justified in requesting that people avoid running up to me, or anyone else, when alone at night. The worst case scenario is really bad, and I have no way of avoiding it. The result of that encounter has little to do with me.
Be honest with yourself, do you feel no apprehension in those settings? Do you not feel that people should generally avoid placing you in that situation?
Or to put it another way once you frame the question correctly and have a bit of relevant data, it's a foregone conclusion that concern for one's personal safety after a polite sexual proposition in a hotel lift at a conference is unwarranted?
This is actually a comment directed at a different issue. It has to do with the search for stories of sexism that will obviously not yield any under the chosen method. There are easily available means of obtaining that information, but that's not being done.