The Atlantic - Latest Comments in Close your eyes, and hold your breath...

Re: Close your eyes, and hold your breath...

metricpenny — Sun, 11 Jan 2009 15:06:10 -0000

Good luck with your research.

If you're successful it will not change the fact that a segment of America (GLBT) was hurt by unfair treatment (the passage of Prop 8) and lashed out against an easy target (black CA voters).

You have Andrew Sullivan's ear and he has a huge soapbox. How about bending his ear and encouraging him to change his approach?

His approach reminds me of that of the early Malcolm X's to civil rights - lash out at those you perceive to be against your cause.

The GLBT rights movement needs a Dr. King.

Re: Close your eyes, and hold your breath...

RhondaCoca — Sun, 11 Jan 2009 01:35:20 -0000

Dark Rose,

I voted on not one proposition on the ballots out here in NY. I skipped over them. Is that possible on a Cali ballot? Does anyone know?

Also TNC, you said this:

"Awhile back you talked about the problem with telling black kids that there are more of us in jail than in college. This is like that. It's not that we don't think there's an achievement gap problem. It's not that we don't want brothers to succeed at the highest level. But scare tactics and exaggeration don't help in that fight. They hinder. It's the same thing here. We need the science on our side. And then we will win."

Exactly, I say this all of the time.

Re: Close your eyes, and hold your breath...

statsman — Sat, 10 Jan 2009 23:08:33 -0000

Re: Table 3 of the report using precinct-level election returns.

It appears to me that it might be a more useful exercise to smooth the data points only for precincts with >50% of black voters. We know that the five cities from which data were drawn are quite racially segregated (see racial segregation indexes). We also know that most blacks in California live in majority black neighborhoods (cites are found pretty easily in a google scholar search). If we assume that blacks in segregated neighborhoods vote at the same rate as blacks in integrated neighborhoods we also assume that most black voters live in majority black neighborhoods. The majority of blacks who casted a ballot then are in the right half of that squiggly line graph. When we eyeball the percentage voting yes for that group the number is much higher than the aggregate prediction.

Re: Close your eyes, and hold your breath...

Darkrose — Fri, 09 Jan 2009 20:28:00 -0000

sgwhiteinflaL I am being real here in that I would bet you that many black folks came out to vote for Obama and weren't pressed about everything else.

Good point--especially when you consider that a lot of CA voters ignore the ballot measures because they're so damn many of them. Our household gets two copies of the voters' guide, and if those are in the mailbox, nothing else will fit. And I read through the whole thing this year, and I'm still not 100% sure exactly what all of them were about.

Re: Close your eyes, and hold your breath...

someone — Fri, 09 Jan 2009 16:00:00 -0000

This is admirable, but be careful not to favor the "facts" that confirm your pre-fact finding beliefs.

Re: Close your eyes, and hold your breath...

Timothy Kincaid — Fri, 09 Jan 2009 15:28:39 -0000

Coates,

I look forward to what your communication with the pros yields.

Re: Close your eyes, and hold your breath...

dlp — Fri, 09 Jan 2009 14:13:28 -0000

Ive done a lot of work on surveys (Im a phd candidate with an MA who works professionally with surveys), and I would like to compliment you on trying to talk to people who know about survey methodology on this one.

Unlike what most people think, a lot of statistics and survey methodology is counter intuitive, or at least complex enough that pure intuition is not enough.

As to Kincaid's points:

The running mean smoother vs arithmetic mean comment in completely off-base, as a running mean smoother is an acceptable, if incomplete, way of getting at the trends in the data. An arithmetic mean basically "flatens" the relationship, making it linear and without change. By using the smoother, you are able to show both the non linear (i.e., it goes down before it goes up) aspect and how much it varies with the proportions of the independent variable. So that part of the "critique" is completely unfounded.

The second key mistake is one that anyone paying attention can spot: when he claims that, because the difference in support for prop 8 is smaller between religious and non religious blacks than in other groups, that religion played less of a role. That is simply innacurate, and does not refute at all the point being raised. As long as there is SOME difference (and 18 points is quite a difference), than the relative proportion of each group within the group as a whole is indeed quite significant.

But the most glaring, and obvious mistake, and also the one that the original report address, but Kincaid does not, is that an identity is not defined solely by race. Education, class, religion, gender, etc. all matter. Kincaid conveniently ignores the multivariate analysis on pages 5 and 7, where these competing variables are looked at simultaneously, and goes back to the same issue as before.

The report's analysis, regardless of what Kincaid thinks of the organization, is technically "good enough." It is not perfect because it does not include certain key variables which are missing, such as education and income. I bet that if/when someone does a study including income and education on the multivariate analysis, the "race" coefficient will reduce even more.

I cant believe people, including Sullivan, are taking this tool seriously. Although commissioned by that organization, the report was not written by them, but by two PhDs in respected universities, who would certainly be looked down upon if they wrote shoddy statistics. And yet Kincaid feels like, without cracking a book on statistics or even googling what some of the terms mean, and ignoring completely the key tables, he knows enough to refute it in its entirety because of how he feels, and what the graphs "look like."

Pure garbage from someone who found what he wanted to find. It is one thing when you feel a certain way to go out and try to honestly educate yourself about the subject on which you want to comment on to see if your feelings are supported or not. It is another thing entirely to nitpick results, critique what you dont understand, ignore what is not convenient, and then put your foot down and claim victory. What Kincaid did was the latter. Willingly ignoring oneself's own ignorance is not a way to conduct an honest debate.

Re: Close your eyes, and hold your breath...

sv — Fri, 09 Jan 2009 14:06:55 -0000

PS keep in mind that there's only so much useful data to be had; you will probably get a rather incomplete picture of how people feel about this.

Re: Close your eyes, and hold your breath...

sv — Fri, 09 Jan 2009 14:03:37 -0000

Coates,

I think that's the right approach, although as someone in the science/tech area I'm "biased towards the facts". Do that reporter thing, dig deeper.

I will say that while it's important to combat black homophobia or anti-gay bias with facts, it seems to be getting a disproportionate amount of attention. Sully quoted someone who had it right, some gay rights advocate saying basically, 'Look, whatever the figures, we need to do more outreach into the black community going forward, in addition to the general public, because that's an area we've neglected and it's simply the right thing to do'. right on.

I also feel you on feeling embarrassed by criminals of your own race. It may not be fair, but they really do sort of make your people look bad. When the Blago thing broke, two of the first guys picked up for allegedly trying to arrange a 'senate seat sale' were an Indian businessman and an Indian Blago aide, I believe. I was totally embarrassed and not a little pissed off. "This fucking makes us look bad," I said. I mean if I can be proud of a Surgeon General Gupta and a Governor Jindal, shouldn't I also be ashamed of Indian crooks? They shouldn't logically represent who we are to the wider community (and thankfully for us, they don't) but that's not the kind of publicity we need.

Re: Close your eyes, and hold your breath...

mark — Fri, 09 Jan 2009 11:29:03 -0000

Second what Nitpicker said:

Nate Silver at www.fivethirtyeight.com does a ton of statistical analysis on polls. (coming from his baseball statistical analysis background he has learned not to let his personal feelings get in the way of the analysis)

He's certainly worth making inquiries to.

Re: Close your eyes, and hold your breath...

Andrew Fly — Fri, 09 Jan 2009 11:14:28 -0000

that analysis you linked too is murky at best But that just doesn’t make any mathematical sense. In their Table 1, they lay out their breakdown of ethnic voting:

(68%) White 49

(7%) Black 58

(14%) Latino 59

(7%) Asian 48

Well sorry, but those numbers don’t get us to 52.3% support. One of those ethnic demographics is understated.

Um, of course that doesn't add up to 52.3 percent, because the sample's percentages (at left) only add up to 96%. There's 4 percent of others that is unaccounted for.

Let's say that that 4% voted 52.3% for prop 8, that gets the total to 51.1. Which could be the result of a rounding error.

Also, let's look at what would have happened if individual groups voted at a ridiculously low rate of 45% yes.

If Blacks had voted 45% YES, lower than any other group, it would have passed with 51.4. If latinos voted 45% YES, it would have passed with 50.5. If whites had voted 45% YES, it would have failed, at 49.6%. It's probably way more cost effective to sway white voters. But blame the blacks, oh noes!

Re: Close your eyes, and hold your breath...

sgwhiteinfla — Fri, 09 Jan 2009 11:10:46 -0000

Coates

I feel you on that. I have tried to defend or dismiss the numbers on Prop 8 also so I absolutely feel you on your analogy. Keep fighting the good fight.

Re: Close your eyes, and hold your breath...

Ta-Nehisi Coates — Fri, 09 Jan 2009 11:05:47 -0000

I'm don't totally understand what it means that blacks who attended church weekly, were less likely to support Prop 8, than other ethnic groups with the same pattern. Whereas blacks who went to church less than once a week, were MORE likely to vote against Prop 8. The analysis Andrew links to argues that if that's the case, then we'd have to say religion isn't much of factor. My old reporter's intuition says that there's something wrong with that conclusion. But I don't really have the IQ to explain to you what it is. I need to talk to some people who know more. Review it yourself, if you like.

I think what's driving me here is that, as I've said before, I'm embarrassed--even at the 58 percent mark, I'm embarrassed. Maybe that's not right, but it is how I feel, and that's all I have. It's like when your watching CNN and you hear about a crime, and your pray the perp isn't black. Now it may be true that you shouldn't carry that. But we all do, don't we? Deep inside we want our folks to both suceed and do the right thing. It's a flawed impulse, but an impulse nonetheless.

So how do we get there? Well my job is to make sure the convo doesn't die. But I need to believe in my information. There are people in my life who I'm having the gay marriage argument with, as we speak. I need to have my facts in order. I need to have my numbers straight.

Awhile back you talked about the problem with telling black kids that there are more of us in jail than in college. This is like that. It's not that we don't think there's an achievement gap problem. It's not that we don't want brothers to succeed at the highest level. But scare tactics and exaggeration don't help in that fight. They hinder. It's the same thing here. We need the science on our side. And then we will win.

Re: Close your eyes, and hold your breath...

sgwhiteinfla — Fri, 09 Jan 2009 10:53:01 -0000

Coates

Ok I got you on that and like I said above I wasn't trying to say what your motivation was. I just wanted you to take a breath on it for a moment and examine IF your motivations might have been misplaced.

BTW what do you think you might have missed?

Re: Close your eyes, and hold your breath...

Nitpicker — Fri, 09 Jan 2009 10:52:17 -0000

Nate Silver at www.fivethirtyeight.com seems to know a lot about polls, especially exit polls. What about the two of you working together?

Re: Close your eyes, and hold your breath...

Ta-Nehisi Coates — Fri, 09 Jan 2009 10:50:29 -0000

Sg,

You're looking at it wrong. This is as much about my own curiosity, as it is about winning. Maybe I missed something. Blogging, for me, is as much about thinking out loud, as it is about the actual debate. This is part of refining my own thinking.

Re: Close your eyes, and hold your breath...

MikeF — Fri, 09 Jan 2009 10:46:58 -0000

One thing you might want to examine is whether the backlash against blacks voting "Yes" could have affected the polling data: JFK was elected by a very narrow margin, but after his assassination something like 65% of Americans reported that they had voted for him. I suspect that this effect only applies to events in the somewhat-distant past, but it could be worth checking out.

Re: Close your eyes, and hold your breath...

sgwhiteinfla — Fri, 09 Jan 2009 10:46:15 -0000

Coates

Seriously man I am with you on getting the facts straight but I honestly think this is a flame war you can't win. Especially with someone as obstinate as Sully. We will never have totally accurate numbers on the racial breakdown of how blacks voted on Prop 8. And because that leaves a gap where either side can editorialize every time you find a study that proves your point he will find one that proves his. Doesn't matter if any of them are total bullshit or not. And I don't mean this in a negative way at all but the fact that he is gay and you are not is going to lend him more credibility from the public at large on this subject most of the time.

Here is what I think is more productive. Take the raw numbers of black voters who came out and voted on Nov 4th. Then play with the percentages to see how many of those people would have had to vote for Prop 8 to have made a significant difference. That should tell you more than this back and forth with polls because honestly I think that even if 70 percent of black folks did vote for Prop 8 I don't think it caused it to pass. Bigger than that how many of those voting even knew what Prop 8 was all about. I am being real here in that I would bet you that many black folks came out to vote for Obama and weren't pressed about everything else.

Homophobia is a problem in our culture and even you have acknowledged that. So if your argument isn't just about if we were the deciding force in passing it stop yourself for a moment and ask yourself what really is your argument? I think when you sit back and think about it trying to prove that we voted 12 percent less for Prop 8 than was reported solely for the sake of giving us a lower number really is trivial. I am only making this suggestion and I am not trying to say that IS your reason behind all this but dude if it is LIG.

Re: Close your eyes, and hold your breath...

DaveinHackensack — Fri, 09 Jan 2009 10:33:17 -0000

"But I learned something the other day from that OED debate--I need to go do my job. Yesterday, I started making calls to do some reporting on the new report, on the original exit polls and the critique of both."

Props to you for using your platform to do some reporting. I've long wondered why some of your Atlantic colleagues didn't take advantage of that.

Re: Close your eyes, and hold your breath...

KevDog — Fri, 09 Jan 2009 10:04:01 -0000

I'm just guessing, but I bet that when you say "This is Ta-Nehisi Coates" you get much higher level of cooperation when you add "from The Atlantic".

Use that platform, man. I'm curious to see what you find.