It doesn’t matter – this isn’t necessary for there to be an effect on test scores. I don’t know if there’s evidence of this, but I’m pretty confident that there’s evidence (like posted above by Evil Economist) that such discrimination can affect test performance.
That may be so. But you have no way to assess how likely it is, which is what’s relevant to the specific question you asked (post #1115).
Until you can come up with some sort of realistic assessment of the impact of the level of day to day discrimination faced by black Americans, you can’t assess the likelihood that this, rather than any sort of genetic factor, accounts for the current disparity. (Which is what I said in post #1116.)
I can compare their likelihoods because there is direct evidence in refutation of one of the explanations – genetics. There is no direct evidence in refutation of day-to-day discrimination as a cause.
As you said in post #1096, we can compare their “theoretical basis” – and both genetics and discrimination are a reasonable cause, with a reasonable theoretical basis. Since there is direct evidence refuting genes but not discrimination, we can conclude that, at this time, discrimination is a stronger and more likely cause than genes.
That’s not how it works. The a priori likelihood is too big a component to just ignore. If you don’t know it, then you can’t make any sort of likelihood assessment.
You’re repeating yourself here, as am I in response. Unless you have anything new on this, I have nothing further to add.
This makes absolutely no sense. We have two explanations (among many others) – genes and discrimination. They both have reasonable/comparable theoretical basis. There is no direct evidence that points specifically at either explanation. There is direct evidence that points specifically away from genes, but not from discrimination. Therefore, discrimination is a stronger/more likely explanation.
These seems pretty simple. Which part of that chain are you challenging?
The word “comparable”.
Then we disagree on this point. What stronger theoretical basis is there for genes than for discrimination?
I don’t know that it’s stronger. But I also don’t know that it’s comparable. They are both unknown.
The lover of Bayesian statistics should be aware that we can never know the a priori liklihood for any social sciences statistic. So if you claim that you can’t make any comparisons until you know the a prioris, then you are saying that you can never know anything.
I’m not sure that’s true across the board, but even if it is, that’s at most only true of precise measures. That doesn’t mean you can’t get a sense.
But more important than that: no one is saying you can’t know anything without knowing the a priori likelihood. What I’m saying is that in a situation like this one, where there’s a paucity of actual evidence for either side, the a priori assumption weighs heavily.
The point is that iiandyiiii is saying that since both alternatives are theoretically plausible, the mere fact that there is some evidence against one explanation and not the other means that the second is more likely. This is incorrect, since the strength of the counter-evidence needs to be balanced against the relative a priori likelihoods of both alternatives, which are unknown.
That would not apply in a case where the a priori likelihoods were unknown but there was more solid and conclusive evidence for or against one of the alternatives.
How about this as an alternative: If you can’t know the priors, then don’t apply Bayesian statistics?
Or I guess you could just sit around waiting for God to hand you down the priors from on high. Meantime make a decision based on the evidence you do have, which I personally find pretty compelling.
This would be the a priori likelihood that a difficult-to-define trait that has not been linked to any particular gene or set of genes is genetically inherent to a population group that has no possible genetic definition?
Then by this different sort of logic, you would agree that based on the evidence, it’s very foolish to conclude (as CP does) that inferior genes for intelligence among black people are a significant contributor to the test score gap?
Or how about this as an alternative? If you don’t know then just say upfront that you don’t know, instead of just pretending that you do in order to “make a decision”?
That would be the a priori likelihood that some unknown gene or set of genes is more prevalent in some population groups that have no possible genetic definition, and account for part of the current disparity in average measures WRT a certain difficult-to-define trait.
“Inherent” is your strawman term here.
I think it’s wrong. I wouldn’t say it’s foolish.
If you think the evidence for a genetic case is compelling then it wouldn’t violate the logic. I don’t think the evidence is compelling. CP thinks it is. But where he’s going wrong is in overestimating the strength of the evidence, not in the logic.
Then our only disagreement, I suppose, is our vehemence with which we disagree with CP.
So much confusion here. By incorrectly stating the real issues, you create a strawman with all the silly counterarguments to the wrong thing.
Let me help you get by the careful strawman wordsmithing and give you a better recap to bite your teeth into:
-
When we self-identify as a “race,” even though there is not a discrete biological category defining that race, we identify with a group that, on average, has a different gene pool that another self-identified race group. What that means is that, if I self-idenitfy as “black,” the average frequency of genes which drive my phenotypic expressions is different than the average frequency of genes which drive the phenotypic expressions for those who self-identify as “white.” That there is no “discrete biological category” has no bearing on this disparate gene pool whatsoever.
-
“Intelligence” is a word. Various psychometric exams measure the human brain’s ability for reasoning. There are multiple other brain-based skillsets that can be quanitifed, and among them is the ability to perform on academic exams considered to be measures of information retention, intellectual reasoning, and subject mastery. The average performance on those exams–regardless of whether or not they correlate with “intelligence” is vastly different between blacks and the white/asian self-identified group. That disparate performance is not erased by providing a black cohort with every possible advantage such as wealth, income, parental education, and educational opportunity. A similar gap exists across every political system and every cultural and social history across the world. At issue here is not “intelligence” but whether or not the academic skillset gap between blacks and other races is because of a difference in gene pools or nurture.
-
Scales used to measure psychometric evaluations of human neurological function are not perfect, but they are pretty damn good, and those who use them are pretty damn good at how to apply them. However, a skillset such as mathematics is fairly culture-independent, and mathematics is a good example of a skillset where the black/white-asian pattern is stubbornly resistant to being erased. For example, children of educated and wealthy black parents score on par only with children of poverty-stricken and uneducated white parents, and far far below children from white and asian families with equivalent socioeconomic and educational status.
Not knowing the priors is not equivalent to not knowing anything. It’s not possible to know the priors, but I’m not going to throw my hands in the air, say “nothing can be known!” and then cry and eat ice cream–I’m going to make a conclusion based on what *is *known.
Actually, I’d like you to provide me some evidence that you apply this level of semantic nitpicking to any other issue. Do you have a post saying “we can’t say for sure whether this frog is the King of Mexico–unless we know the priors we can’t reach a conclusion one way or the other?” Or are you just trolling this specific topic for some reason?
So I guess this leads to to a clear solution to the issue–have blacks self-identify as East Asian, and watch their test scores go up.
Reread post #1134.
Reread post #1130.
Nothing you say in 1130 is right. Also, you clearly don’t know anything about the topic, so you are not entitled to use the word “paucity”. Try reading post 1118.