I may be prejudiced based on your history in other threads, but I can’t tell if you’re finally conceding that face’s argument is correct.
I have a feeling that the “real-world application” is yet another carefully worded disclaimer. My opinion is that if something doesn’t have application in this world, it’s completely worthless to argue for its merits even in a hypothetical sense. If the logic is faulty here, it’s faulty even in a world full of evil flying rape-monkeys.
Bricker, I don’t think face wants you to don sackcloth and kiss her ass for all eternity. I think she just wishes you would concede that your recent arguments about stats have been incorrect. Not just incorrect here, on planet Earth. Incorrect period. It’s the weasling with the disclaimers that has her riding your case.
Other than the begining opf your post you point out how poor a determining factor the one piece of information is. That is true. But thet question was based on that very limited information, can you make a better guess and a worse guess. And I maintain that answer is yes.
If whites are getting raped 99 times more than blacks are, then yes I’d expect false accusations to track with that rate. Not with the population size. I have no shame in admitting this, for I’m a racist bitch who eats white babies for breakfast.
Hmm, if you want to proudly wear the badge of ‘begbert said I hold racist positions’, shouldn’t you be adhering to the premises I stated? You don’t want to lose your baby-eating licence.
And I suspect that the reason you find my posts, er, ‘cataract-inducing’ is because I have a general personal tendency to use relatively long sentences, which aren’t actually run-on senstences, when I am communicating (and I have a tendency to use parenthetical inserts, as well). Yes, I do this in my everyday speech as well.
And now for something completely different:
It did in all my math, science, and economics classes. When working with a hypothetical scenario, you start with your stated premises/asumptions, and that’s what you’ve got; everything else is assumed to be irrelevent or (failing that) in the default neutral state.
The biggest problem I had all along with the math was the translation of the english words into an appropriate mathematical model.
The simplest and probably the most accurate conclusion is that both claims are 100% accurate because we haven’t allowed for them to be inaccurate. There is simply a direct translation from event to report of event, no other information is allowed. Clearly this is so trivial it’s not worth discussing.
So we allow for inaccuracies, but we don’t say how or why. We don’t say anything other than “in the absence of other information.” What information is absent? Do we throw out everything we know about the world due to that statement and only allow what was listed in Bricker’s statement? What are we allowed to incorporate back in? That there are white men and women, and black women (no mention of black men so they don’t exist, but if that’s the case then the WoB claims are 100% correct?) At every turn, when you take the text and turn it into a mathematical model you will be making assumptions and decisions based on things you know about the real world, while at the same time taking a select few and setting them to some arbitrary value (population size and rate of claims). begbert2, the math is incorrect not because a figure was multiplied incorrectly or a decimal point was misplaced, it is incorrect because there was a HUGE assumption during the translation from quasi-real world/text/ideas to mathematical model. And that assumption was how to incorporate “all things being equal” and “in the absence of other information” into a model that required some other information.
And if you disagree that “all things being equal” and “in the absence of other information” can actually be ambiguous, despite sounding non-ambiguous, I guarantee you Bricker would have a field day with those statements if they appeared in a contract dispute.
Statistical interpretation is not an art of relativism. There are correct ways of interpreting stats and incorrect ways of interpreting stats. You can choose the incorrect ways, but don’t expect scientists and rational-minded people to respect your conclusions.
As I said, as a fish biologist, I’m tempted to make predictions all the time about where fish are going to be distributed. If I have one piece of information about a given body of water, I don’t make the foolhardy mistake of making a guess from that. I either go out and collect more information, or throw out that site from my analysis. If someone was holding a gun to head with the safety off, then maybe I’d make a guess. But that doesn’t mean my guess will be in any way accurate. I’m so very happy that I would never find myself in a situation where someone will kill me if I can’t tell them how many fish they’re going to catch!
I maintain that it is easier and wiser to withhold judgement in the face of limited information rather than make WAGs.
When you are working with data collected from human beings, you can not make assumptions that amount to leaps of faith. The fewer and smaller the assumptions made, the better. Epidemiology–the study of diseases within populations–uses the same tools as economics, but a completely different kind of thinking.
For a mathematician, making numbers add up to the right thing is key. For someone trained to look at animal herds (whether those animals are human or not), assuming the right thing is far more important. Because if your assumptions are baseless and arbitrary, then whatever digits you come up will amount to shit. “All things being equal” is one of these baseless assumptions, because few things are equal, including populations. It makes little sense to say “all things being equal” and then make one group subject to rape 99 times more than the other, just to make some useless point about what happen if this assumption was true.
I may well have missed it, and I certainly agree with it. It’s an extremely unwise and useless move, and I certainly never contended otherwise. My point all along has been a technical one.
You’re right – there is no rush. And since I always conceded that REAL evidence would trump the tiny scintilla adduced by statistics, I agree in the main with your point. But I never claimed otherwise. I answered a technical point.
No harm done. I’m sorry myself for simply fueling the aggressive fires rather than explaining my point without compounding it with insults. Perhaps a good lesson for this is: even if it’s in the Pit, it doesn’t HAVE TO contain insults.
It looks like a cease-fire may be in the works, gang.
[/QUOTE]
Can you clarify something for me then? In post 335 of the Heurta thread, you wrote this:
What is so bad about declaring racial crime statistics irrelevant in the middle of a discussion about the Duke case? I don’t get why such a thing is such a horrible offense and warrants calling me stupid, especially if you agreed with my main point, as you say now. You might have noticed that Hentor stated that the stats in question were irrelevant, too. They are irrelevant precisely because they “don’t say what you claim they say”. For dozens of pages I spelled out why. It beggars belief that you would have missed what I wrote, especially when EE was kind enough to post some of the arguments I made.
I really want to change my impression of you as a poster. Really I do. But you still keep insisting that “you were answering a technical point”. When you say stuff like this:
(bolding mine)
It is hard to trust what you have to say. If this all is a great big misunderstanding, I don’t see how you could have praised Heurta like this. You may have been talking theory, but he definitely was not. In fact, he was spouting his probablistic model crap even though by that time there was plenty of evidence.
I want a cease-fire, but it still seems like you are evading responsibility for things you’ve written. Can you at least concede that you wrong about Huerta and that I had every reason to find his position troubling, with or without his bullshitty disclaimers?
We agree. Absolutely. But that doesn’t stop me from being able to suspend reality and entertain a hypothetical. Even if I’m right and we can make a better guess/worse guess in the hypothetical, that desn’t mean that I would act accordingly in the real world. For instance, in the real world you would immediatley be barraged with more, and more reliable, information: the person’s demeanor, the time of day, location, whether or not they had made a false claim before, whether or not there were witnesses, etc.
It’s unfortunate that you didn’t focus on my sentence in this way:
Huerta88 backs up his posts with cites, and argues reasonable inferences from those cites. I’m not sure I’m completely on-board with everything he’s said, but he’s made a very good case for himself.
In that sentence, I was prasing his METHOD of presenting his case. You were angry - furious, even - and it showed. Huerta was not. He posted calmly, offering cites. You did not. I was not on board with the conclusions he was drawing – hence, my statement I’m not sure I’m completely on-board with everything he’s said. But in the discussion between the two of you, he came across as a reasonable, dispassionate rhetor, and you came across as an angry yeller.
As I said all along, Huerta’s position, itself, was not solidly grounded. But in that thread, his presentation was more convincing.
I should have taken more time to make that distinction clear, but my own ire got… er… ired up, and I began throwing insults myself, which of course made my own efforts to explain things exponentially less effective. Physician, heal thyself, as it were.
It is well to remember that “one may smile and smile and be a villain.” It takes superhuman strength not to get increasingly frustrated when others look adoringly at Eddie Haskell for his obsequiousness, while you know what a little shit he really is.
That’s a different statement than “I’m not on board with everything he’s said.” What you actually said made it sound like you were largely in agreement with only some small degree of difference.
I admit to being more irate about the topic than I would otherwise be, but that was because I just can’t understand holding Huerta88’s position without the underlying supposition that one claim is implausible because, really, why would a white man want to have sex with a black woman. No, I admit that is not a fair position on my part, but without a logical substrate, it is hard to understand why someone would believe that a black woman’s claim about interracial rape would be less credible.
It was, I admit, reinforced when I went to check the data upon which his dubious and illogical claim even rested in the first place, and saw that he cherry picked one years worth of data that fit his assertion and failed to mention several other years of data that did not.
I wish that we could discuss the issue of using probabilities from other samples to make a decision between the likelihood of two new individuals’ claims, outside the topic of interracial rape. I see the logic employed by those on the other side, but I find it erroneously applied. Yet I also agree that at the margins, given high degrees of implausibility (e.g. dead Lincolns and editor monkeys), some credulity is strained. So, is there a line? Is there any way to determine a line? These would be interesting questions.
Hentor that is IMO a fair take on what was going on in that thread and that’s also my take why there’s this, I don’t want to discuss that, lets talk about ninjas vs dragons.
There’s been attempts to frame the question in ‘generic’ terms, but even those questions are never answered instead they are modified to flipping nickles or Alec Baldwin licking some guy’s sack.
Here I’ll try again. Here’s what is known:
Redheaded people are about 4% of the population in this town. The remainder have brown hair, that’s 96%
Farmer Frank has ten sheep. Seven are white, the other three are black. He left the gate open and the sheep escaped.
Mr. Jones has red hair and says that one of Farmer Frank’s black sheep is in his garden.
Mr. Smith has brown hair and says that one of Farmer Frank’s black sheep is his garden.
Mr. Smith and Mr. Jones are both regarded as equally honest.
Question A: is this a fair representation of the question we’re trying to ask? If not please feel free to modify it, but remember the question isn’t to chose which one is lying. It’s to explain why one’s claim is more credible, than the other. That’s it.
Question 2: Is it enough to say Mr. Jones less credible than Mr. Smith because he has Red Hair and is part of a minority group which due to it’s representation, comes into less contact with “stuff”, than the majority?
Would it make sense for me to say that based on the number of Red Heads in a population, all things being equal; red headed people are less credible, people with brown hair…even to the .000001 percent?
Redheaded people are about 4% of the population in this town. The remainder have brown hair, that’s 96%
Farmer Frank has ten sheep. Eight are white, the other two are black. He left the gate open and the sheep escaped.
Mr. Jones has red hair and says that one of Farmer Frank’s black sheep is in his garden.
Mr. Smith has brown hair and says that one of Farmer Frank’s white sheep is his garden.
Mr. Smith and Mr. Jones are both regarded as equally honest.
I have one particular problem with your hypothetical, in terms of approximating the parameters of the other problem, but first, you should specify that people with brown hair and people with red hair are equally honest, and not specifically these two people.
But I think the larger problem with your hypothetical, in my opinion, is that the set of sheep and their color distribution is a known quantity. That is, the sheep in question are known to have come from a group of 10, with three black ones. I think your hypothetical makes the point, but it doesn’t resolve the inherent problems with extending frequencies within a sample to probabilities regarding elements of another set.
I think the confusion others might have been having is that, yes, if you have a bag with 75 blue marbles and 25 red ones, you can fix a probability of reaching into the bag and pulling out a blue marble. But if I walk up to you having just come from the toy store with a marble in each hand, what’s the probability that the one in my left hand is blue and the one in my right hand is red? You just can’t assume that your proportions will be invariant with those of the sample from which I drew mine.
Similarly, if you had a room full of black and white women claiming rape AND knew how many were lying, you could fix a probability that you would randomly draw out a white liar or black truth teller. But two new women coming from outside the room and claiming rape - what are the distributions within the sample that they came from? It would be foolish to think that rates of rape claims will not vary over time and across samples.
That said, I like your hypothetical for simplicity’s sake, and don’t mean to poop on it. Besides, who really wants to have sex with a black sheep, anyways?
Yes, but probability isn’t the same as credibility. If i know you have blue marbles and red marbles and you say you picked a red one; it doesn’t matter whether you have 1 red marble or 99; all that matters is whether or not you could have picked a red marble. If the answer is yes, then that should be good enough…if I have no reason to doubt your credibilty.
I think you’ve done a great job at it, fwiw. and my answer would be: since both events are** possible**, even if they’re of varying probabilities, the farmers would be equally credible, that is I would have no reason to doubt their statement even though in the world one of the events is more likely to have happened than the other.
and I think maybe that’s the crux of the biscuit as it were.
I agree entirely. I’m just pointing out further flaws in the argument that one might apply frequencies in the manner that has been done so here. Having said that, there’s only one line from your analogy that will be necessary to answer the question.
One problem with this, as it relates to the original 2 statements we have debated, is that Bricker did not imply the color of the skin of the person making the claim was a factor, merely the color of the skin of the two individuals involved in the event.
In other words, if both claims were made by a third party (hispanic, computer, anonymous tip, etc.), Bricker would still have applied the same reasoning to the 2 claims being made purely based on the statistics regarding the races involved in the event.
So in this case, I think it’s an error to introduce the red hair into the final equation, and I think Bricker would agree.
Don’t get me wrong, I still have a problem with assigning the credibility of a report based on the frequency of the event. I just don’t think Bricker was intending the skin color of the person filing the report to be a factor beyond the fact that the filer was also the person involved in the event.