It’s not mutually exclusive, man. I dig men. Blacks, whites, Latinos, Asians, give me them all.
No it isn’t. There is no one “black look” and there is no one “white look”. There are black people with skin as light as milk. There are black people with blonde-hair and blue eyes. There are white people with brown complexions and tightly curly hair.
Appearance, just like culture and other things, is associated with race but not dictated by it.
Maybe it would. Maybe it wouldn’t. It wasn’t a factor with the men I’ve dated. So I can’t assume that it is for you without knowing more about you.
The racial preference of an “average” white rapist is not relevant to the Duke allegation.
For a number of reasons:
The accused are not “average” white guys. They are individuals who happen to be white, with their own histories, socioeconomic backgrounds, and likes and dislikes. I’m sure that income alone makes these guys very “unaverage”.
They are not accused of a random act of rape. They allegedly invited the strippers to their party, which means they saw the accuser as a sexual object. Perhaps if these guys were cruising town, scouting out chicks to score with, then racial preference would be relevant.
There’s no evidence that this rape–if it happened–was due to sexual attraction. If the guys were spouting out racial invectives as the alleged, they may have wanted to escalate their hatred through humiliation of the accuser. Only the rapist knows why he rapes, and since these guys deny raping this girl, we’re not going to know why they did what they allegedly did.
Giving these kinds of stats relevance means we assume that white people are a fungible group. In the US, “whites” include everyone from Europe and the Middle-East. What if Middle-Eastern men and black women have high rates of romantic pairings? This will not show up in data drawn across all “white” ethnic groups, and we will wrongly conclude that a Iranian has the same racial preference as a German.
If the accused were Egyptian, do you think Huerta would have brought up the DOJ statistics?
That’s the fundamental point that seems to get lost in this discussion about statistics. It’s not a cse of getting the arithmatic right, it’s more a matter of whether some national data on the relative occurance of interracial rape can meaningfully tell us anything worthwhile about this case at hand. But the fact is, we realy don’t know exactly how rapists choose their victims. We observe some general trends, but this case has many peculiar characteristics. We’re not talking about some woman claiming to be raped as she was walking to her car at night.
And even if we could use some type of DoJ statistics, which ones would we use:
The average of all rapes in the US?
The average of rapes in North Carolina?
The average of rapes in the county where Duke is?
The average of all counties in the US with similar demographics?
The average of rapes in the age range of the victim and the accused?
The average for college students?
The average for strippers?
The average for strippers being raped at frat parties?
With all the things going on that night, why we should focus only on the race of the individuals is just beyond me. Maybe it’s because one of the first thing many people tended to notice (or that the newspapers reported) was that the woman was Black and the guys were White. But just becaue that was noticed first doesn’t mean it was of primary importance-- in fact it doesn’t mean anyting at all.
Blacks are more likely to rape other blacks than whites. If a white woman accuses a black guy of raping her, will people cite crime stats to support the accused guy’s case?
In my answer to Dragon Ash’s question about “average” racial preference being relevant, I forgot another point (one that YWTF mentioned).
Black people span a spectrum of looks. Some look white (like Vanessa Williams). Many are considered beautiful by everyone’s standards. They, like white guys, are individuals with their own histories and sets of behavior.
I doubt seriously people would be talking about racial preference if we knew the accuser in this case looks like Halle Berry.
Sure you’ve ducked something. And now you’re trying to further obfuscate. Your formulas, intended to explain something to me, are identical to and yield the same answers as mine, although you for some reason neglected to calculate the second one. Both calculations yield .95, of course, which I think anyone with half a brain would realize from the hypothetical. Is your point that the hypothetical is not real data with real sensivity and specificity? Not much of a point. Malacandra explicitly stated that he was operating in the “frictionless” world of hypotheticals, and challenged you to make a simple calculation based on the facts presented. You ducked. And you’re still ducking.
I’ll probably regret stepping into this mess, but:
What he’s talking about is the predictive value of the positive test.
We already know the sensitivity (presumably, the 95% ID rate).
But even with an extremely sensitive test, false positives outweigh false negatives in populations with an extremely low disease/condition rate. That’s why, for instance, screening the general population for HIV wasn’t considered a useful strategy.
So in this case, even though the sensitivity of the maroon-detector is high, the rarity of maroon cars means that the predictive value of a positive “maroon” result is relatively low.
Those (positive and negative predictive value) are the other two boxes in the Bayesian square.
But you sure aren’t stepping up to challenge what’s she saying, are you? If what she has said is so wrong, then you should be able to dissect her argument and tell the audience just what’s so wrong with it. I guess it’s easier to sit on the sidelines and chuckle rather than actually enter the fight.
I bet with **John Mace[/] now in the thread you’re not going to keep rattling off about the relevance of stats to this case. I dare you to.
pazu, I think that Malacandra was simply asking you with the face to demonstrate an ability to perform simple statistical calculations. She responded by blowing smoke.
I predict that having Pazu post in agreement with you with the face, the idiots who are fighting with her will quiet down. It seems to me that the fools in this thread automatically disagree with her no matter what she says. Anyone else noticed this?
Other posters have jumped on you with the face’s bandwagon, and yet she is the one who’s said not to know what she’s talking about.
YWTF, if you had never let anyone know you were a black woman, I think you would be getting a fairer shake around here. I know saying this is only adding fuel to an inferno (to which I say “Bring it on”), but I can’t think of any other reason why posters are giving you such a frickin’ impossible time here. And who can blame me for thinking so, based on the obsession with race evidenced in this thread?
That’s okay; I failed to demonstrate an ability to type proper English in my post:
But even with an extremely sensitive test, false positives outweigh true positives in populations with an extremely low disease/condition rate. That’s why, for instance, screening the general population for HIV wasn’t considered a useful strategy.
It still appears to me that he presented a simple math problem for her to solve. His intent may have been to spring a big “gotcha now” on her when she solved it correctly, but it seems to me that he was trying to cast doubt on her statistical expertise.
Huh? That’s not true at all. These were your forumlas.
That’s nothing like this:
Notice how my terms are different between the denominators, while yours are not.
Before you define what a “true negative” is you need to define the question that’s being tested. On the basis of the hypothetical’s set up, that question reads “How accurately can this person identify a maroon car from a population from reds and maroon?”
If the test was designed to show how accurately the person can ID red cars, the “needle in the haystack” would be reds. But it’s not, since the witness is being exposed to 49 times more reds than maroons. Are you following me up to this point?
My point is that the 95% stat only applies to specificity but gives us no clue about the sensitity. As you can see from my formulas, the terms required to calculate sensitivity are unknown based on the info in the hypothetical. We don’t know how many maroons were correctly ID’d, and we don’t know how many were IDd incorrectly.
*We do know this term, my bad. It should be 49. Got carried away with the question marks.
Crotalus, Malacandra and YWTF herself are doing a fine job of ripping to shreads the idea that YWTF has any clue as to even simple mathmatical calculations. Here’s a tip to pass on to your sister: Everyone makes mistakes. When you make one as obvious as the one she made with the red/maroon car question, the correct response is NOT to despertly continue to blow smoke and obfuscate, it’s to say “Hey, you know what? I bungled that one. I was wrong. Sorry about that” and move on. If she’d done that it would have been over. Instead she has chosen to take the route of the holocaust deniers, despertly denying all evidence and looking like even more of a fool because of it.
And what does John Mace have to do with any of this? The stats remain relative as I have always stated they were: As one data point to be factored in when considering how likely the accuser’s story is to be true. Not as a determiner of her guilt or innocence, not as the sole data point, not to be used in contradiction to actual evidence, not to determine absolutely if her story is true or not, but simply as one data point indicating how likely her story is to be true. Not all of your bombast and hysteria will change that, nor will other facts about this particular case. IF someone produces an unedited videotape tomorrow showing the accuser in this particular case being raped by the entire Duke lax team, Big Bird and Fidel Castro, while the Count Basie Orchestra plays Stardust in accompanyment, and Jesus Christ desends from heaven to personally testify that he witnessed those events, it still won’t change the fact that the crime she is claiming occured is statistically unlikely. It will just make the next one statistically more likely by a tiny amount.
That’s not sensitivity. That’s specificity. As I’ve pointed out already, we know this simply on the basis of the test design and the way that Malacandra worded the 95% statistic.
Identifying a red car correctly 95% is easy if there are 980 red cars in the test. It’s obvious that you could guess and be right most of the time. When in doubt, say red and odds are you’ll be right.
What we don’t know is how easy for the person to identify the 20 maroons. If they bat 100% on all the maroons, then it means very little if they slip up a few times on the reds. The witness is only credible if they can demonstrate skill in finding the needle in the haystack.
You know, I’ve been pretty civil with you. I have no idea why you continue diss me and say I’m blowing smoke when I’ve taken pains to explain everything that is wrong with Malandra’s conclusions and yours.