Here’s another one…

If a white person and a black person commit the same crime, the black person is six time more likely to be featured on the news. If AO doesn’t think that colors the judgment of future juries, well, then he truly is a Republican.

Speaking of which, it is clear to me if AO was on the jury in two cases with the exact same evidence, he would vote guilty to convict a black man and vote innocent for the white man.

AO,

Yes, it does give me pause for thought. And that’s the point I (and I think others) was trying to make. Why are minorities more likely to be involved in crime and what do you propose we do about it? Let’s say I conduct your experiment, and I get mugged 10 times more often by minorities than Whites. What then? It’s important to note these figures, but as I said, there’s more to research than just presenting raw results (thought that’s the important first step).

Southern quaker,
You might be modestly alarmed but whites are roughly 70% of the population so the numbers don’t seem that far off. What might give you pause for thought is black men are roughly 7% of the population yet they commit over 50% of the murders.

Hick,
Ask a taxi driver or better yet make the experiment that was suggested to Quaker and flash a roll of bills around different neighborhoods and get back to us with the results.

Nearly 70% of those serving time for violent crimes against children are white males. In fact, it seems likely to me, based on these statistics, that crimes against children should correlate with the percentage of white men in a community. Ergo, I should be highly concerned for the safety of my children if a white male moves in next door, right?

Suppose you flip a coin five times and it comes up heads four times. What prediction can you make about the next flip of the coin?

Here’s a hint: the individual instance is not affected by previous results.

I have. You must have already gone nighty-night when Mr. Renn showed up. Turns out you’re wrong in ways I hadn’t even imagined.

In this case Quaker it is an illustration of a concept you choose to not understand. Probability.

Probability does not apply to the behavior of individual members in a sample population. Especially not a sample population of one.

AO,
As my boss keeps hammering home, results and findings are two different things. In most peer reviewed research articles, the Results section is followed by Discussion and Conclusions section.

Conceding that there is a significant positive bivariate correlation between percent AA or Hispanic in a neighborhood and crime rate, what findings do you draw from this result?

Making judgments about “the chances of one man committing murder” based on his skin color is a textbook example of bigotry

In this case Quaker it is an illustration of a concept you choose to not understand. Probability. In this context abrasive, perhaps. PC, definitely not. But bigoted, we’re talking about probability, based on real numbers. Probability lets the chips fall where they may.

Bald racism seems to be a more apt description for the idea you signed off on that a black man had to be slightly stupid to not opt in to a criminal lifestyle.

Quaker, have you come to grips yet with what the word correlation actually means?

And, by the way, I have been addressing “what was actually said” for two solid days. Making judgments about “the chances of one man committing murder” based on his skin color is a textbook example of bigotry.

If you take this judgment as validation of the idea that “those people” shouldn’t live in our neighborhoods, then it’s bald racism.

Minorities commit crimes out of proportion to their numbers.

AO yesterday:

We can look at two men and knowing nothing but their race we can accurately say that the chances of one man committing murder are greater than another.

Conservative “logic,” ladies and gents.

You bring up the factor of instability. Perhaps you could define that. What metric would you use to measure instability? But let us concede that stability is a factor. And further more let us concede that defining instability and giving it an accurate metric increases the accuracy of predictions built upon probability.

Be that as it may, it changes not the level of accuracy of predictions built upon the probabability of a different factor.

Correlation is not causation, yet it doesn’t change the probabilities based on correlation.

“A person’s race has nothing to do with whether or not that person is more inclined to be a criminal”

I suppose Bill Clinton would say it all depends on what inclined means. However you can state that a person’s race does have something to do with the probability that he is a criminal.

Oliver loves to put conservatives and Republicans in a bad light. In fact it is his day job. There is going to be pushback when he does that. This post of his was a can of worms begging to be thrown back in his face.

Minorities commit crimes out of proportion to their numbers. A sad fact, the country is worse off because of it.

Observer’s central thesis is basically, “When minorities move into the neighborhood, crime goes up.” He wants us to know he really believes that he hasn’t come to that conclusion through any sort of bigotry in his part and is in no way a racist view.

Of course, he purposely omits a part of my paragraph and refuses to answer the question; to wit:

(rockroq) “Anyone who believes that blacks and Hispanics are naturally likely to be of a criminal mind has to be a bigot and racist. Anyone who believes that the act of a non-white person moving into the neighborhood changing the diversity of that neighborhood causes crime to rise is a bigot and a racist. Don’t you agree with that?”

OBSERVER pretends to answer the question by responding to something that was made up in his mind since it’s clear I never said or asked such things:

(OBSERVER): “Crime need not rise with the act of a non white person moving into a neighborhood. Yet if we examine neighborhoods of differing demographic populations will they have identical rates of crime? It is neither bigoted or racist to observe and discuss such differences as may exist.”

Then OBSERVER moves back into position to continue his argument that neighborhood crime will either increase or decrease depending on the color of the peoples’ skin that live in those neighborhoods:

(OBSERVER): “So do you think different communities that can be identified by racial demographics will have identical crime rates?”

No. As I have already said, neighborhood crime will rise in neighborhoods that have been destabilized. Regardless of the racial make up of a neighborhood, the crime rate depends on whether or not that neighborhood is stable.

OBSERVER then goes on to outright fabricate what I have said.

Then OBSERVER asserts: “Whether something is politically correct or not has no bearing on accuracy. To think otherwise is naive at best and creates a fog of ignorance that makes a clear view difficult.”

Let me make this clear: this has nothing to do with being PC. It has everything to do with knowing what causes crime rates to rise in neighborhoods as well as having everything to do with ignorance; which apparently has caused you to believe what you do.

But, see? That’s what happens when you base your beliefs on bumper sticker sayings and simple talking points, OBSERVER.

OBSERVER believes: “It is very hard for someone to defend the position that neighborhoods characterized by different racial demographics have identical crime rates with a straight face.”

To which I say, “RUN WITH THAT!” Make sure that is the message to which conservatives rally to. Make sure you scream it from the mountain tops.

Make sure that everyone knows, when it comes to a human being’s race, conservatives (since our liberal position is flawed and all) believe that minorities moving into the neighborhood alone is going to cause crime rates to increase.

The fact of the matter is that whether or not it is whites, blacks, Hispanics or a majority of any other race that lives in a particular neighborhood, the only decisive factor in whether or not that neighborhood may or may not have a higher crime rate than other neighborhoods is because that neighborhood has become unstable somewhere along the line. It is a socioeconomic factor that will help decide which neighborhoods are more crime ridden than the next one. A person’s race has nothing to do with whether or not that person is more inclined to be a criminal

As to likelihood > null, on the basis of ONE variable (race), no self-respecting researcher would do more than pee on AO’s pet ‘likelihood’ from a great height. Are we talking instead using an ANOVA, or something simpler like a chi-square, or something really complex, like a multiple linear regression? Only the chi-square might lend itself to the single-variable example, but the power of such a result would be tiny, really. Also, we need between- and within-group calculations to explain variance as well, if we are going to have a good r-value.

Still waiting for AO to explain why one family of a different race in a neighborhood would be likely to ‘correlate’ with anything.

Not necessarily. One might pick two variables entirely at random that produce a perfect correlation–a straight line relationship.

Still waiting to see that one.

Main Entry: cor·re·la·tion
Pronunciation: \ˌkȯr-ə-ˈlā-shən, ˌkär-\
Function: noun
Etymology: Medieval Latin correlation-, correlatio, from Latin com- + relation-, relatio relation
Date: 1561
1: the state or relation of being correlated ; specifically : a relation existing between phenomena or things or between mathematical or statistical variables which tend to vary, be associated, or occur together in a way not expected on the basis of chance alone

The basement version is building a specialty of his, a strawman. Either you don’t understand probability or you’re being deliberately argumentative.

As for the one drop rule you mischievously put on the table I will leave that to Oliver. He has already come out in favor of the 50% mark for Obama, he can deal with the finer graduations that also define “black”.

How do mixed race people fit into your interesting theory, AO. If someone has just one black parent are they only half as likely to commit murder?

Or do you subscribe to the one-drop theory?

Tell you what, I’ll provide the definition of correlation for you:
Correlation (often measured as a correlation coefficient) indicates the strength and direction of a linear relationship between two random variables.
(courtesy wiki)

And what was QiB’s original request, again? Oh yeah, it went something like this:If there’s a correlation between race and crime, we should expect there to be a straight-line relationship between percentage of minority population and crime rate, right?

So, um, yeah, I think QiB is okay on the definition of correlation.

As for your own amused ramblings, The idea you are trying to express is that correlation is not causation. Yet correlation alone is enough to make educated guesses of anticipated outcomes with a level of accuracy corresponding to the degree of correlation.

No, it’s not, dimwit. Not unless you control for other possible variables. Like, say, poverty for example.

Maybe the moderator can fix that.