Roberts' questions and conclusion appear to be taken from a census survey cited in a lower court dissent."A lower court dissent" is a funny way to refer to the D.C. Circuit Court of Appeals case that is under review! Roberts pulled something out of the case that the Court is working on. Under the circumstances, it would be bizarre if the Solicitor General didn't get the reference. (Check the transcript PDF at page 32.) Tapping material in the lower court's opinion is predictable and perfectly mundane. Totenberg glosses over that to stress the data underlying the Court of Appeals judge's opinion, which, she tells us, comes from Census Bureau data that have such a wide margin of error that it doesn't really mean much. Well, if that's such an important point, why didn't the Solicitor General say that in the oral argument?! Here's what we got instead:
CHIEF JUSTICE ROBERTS: [D]o you know which State has the worst ratio of white voter turnout to African American voter turnout?Maybe saying "I do not know," Verrilli secretly meant that the Census data was so rough that no one could really "know" such facts, but the transcript shows a blank statement of lack of knowledge and an effort to shift away to the subject of what findings Congress relied on. If the statement in the dissenting opinion (written by Stephen F. Williams) was so unreliable, Verrilli should have shot it down neatly and quickly.
GENERAL VERRILLI: I do not.
CHIEF JUSTICE ROBERTS: Massachusetts. Do you know what has the best, where African American turnout actually exceeds white turnout? Mississippi.
GENERAL VERRILLI: Yes, Mr. Chief Justice. But Congress recognized that expressly in the findings when it reauthorized the act in 2006. It said that the first generation problems had been largely dealt with, but there persisted significant -
CHIEF JUSTICE ROBERTS: Which State has the greatest disparity in registration between white and African American?
GENERAL VERRILLI: I do not know that.
CHIEF JUSTICE ROBERTS: Massachusetts. Third is Mississippi, where again the African American registration rate is higher than the white registration rate.
He didn't. Totenberg is doing cleanup work. She went out and talked to "Census officials" who told her that "these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states." That's useful to know, as the issue in the case has to do with how closely the Voting Right Act tracks the actual problem of voting rights violations in the states.
But "Chief Justice Misconstrued Census Data"?! Why doesn't NPR care about its reputation for journalism? What an embarrassing display of eagerness to discredit Roberts! Totenberg's article isn't about Roberts misconstruing anything. It's about the relatively low value of Census data that Judge Williams used in his dissenting opinion. If that material was so terrible, Verrilli fell short at oral argument.
ADDED: Pepperdine lawprof Derek T. Muller emails noting Totenberg's focus on 2010 census data, when the relevant data — in the Court of Appeals case and for the purposes of the 2006 reenactment — is the 2004 data:
If I may, both Ms. Totenberg and [Massachusetts Secretary of State William F.] Galvin are either intentionally misrepresenting Chief Justice Roberts's (and the lower court's dissenting opinion's) data, or they are unaware of an important distinction they've elided over.Again: embarrassing.
For Chief Justice Roberts (I think), the concern is the coverage formula. And the coverage formula was reauthorized in 2006. And the last available voter data was 2004. It's unsurprising, then, that the lower court's dissenting opinion, at 11-14, look[s] at the voting data from 2004. It specifically refers to this Census data,Table 4a.
Within that table, one can see that the turnout rate for African-Americans in Mississippi in 2004 was 66.8%, MoE 5.2. In Massachusetts, it was 43.5%, MoE 9.6. So assuming one wants to stretch the MoE, the low end of MS would have been 61.6%, and the high end of MA would have been 53.1%. Ms. Totenberg's calculation to "factor in the margins of error at their extremes" would result in the same confidence that MA African-American turnout was worse than MS.
As to the citizen voting-age population question, one can run a quick check in the MA data to see that it would rise from 43.5% to 46.5%, while MS would remain largely the same — and I'm fairly confident that even a change in the MoE would not put MA in a statistical range in which it would be better than MS.
Now, this is important data because *it is 2004 data*, the data that Congress would have used (and, taking into account time and space, absent a DeLorean, *could* have used) when it reauthorized the coverage formula.
Ms. Totenberg and Mr. Galvin use the 2010 Census data, which is not the data that Congress would have had at its disposal in reauthorization.
Mr. Galvin "assumes" it is the 2010 data Chief Justice Roberts discusses, and is not terribly careful if he says the "only thing we could find" was the 2010 Census, or that "academics" at other institutions "could find no record," when the record *is in the lower court dissent itself*.
Ms. Totenberg, to her credit, links to the lower court dissent--but then ignores the actual 2004 Census data cited, instead choosing to cite the 2010 Census data, which was not used in the lower court dissent (and which, I assume, was not cited by Chief Justice Roberts).
Now, granted, I understand that one could argue that the question is too narrow, that citing solely the returns from a single election (i.e., 2004) is not enough to sink the coverage formula, that the effectiveness and turnout rates today are important in the Court's analysis, etc.
But, these stories glibly rejecting a point Chief Justice Roberts made at oral argument by using a point he didn't make do not advance the conversation in any meaningful way.
Posts
95 comments:
Why doesn't NPR care about its reputation for journalism
You know the answer to that. As the great if morose Jeff Goldstein says, It's who they are. It's what they do.
"these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states."
So relying on numbers from the official U.S. Census that is held every ten years is more innaccurate than relying on 50 years old data that was also collected by the U.S. government?
I bet if the census data showed the reverse, it would be all the proof needed to continue the VRA sanctions.
This dismissing the census data because it doens't comport with the leftists political position on is just "heads I win, tails you lose" racist bullshit.
So the governments position is predicated on junk data. Marvelous.
The left has been orchestrating an "anybody who's against extending the VRA is a racist" campaign.
My FB page is filled with alternet and Nation propaganda graphic blasts labeling Scalia as a racist. Now, I guess, Roberts is too.
I tried to discuss this with one of the most virulent commies. We've entered new territory here. "Race" doesn't exist (just a social construct, whatever that is), but "racism" does. I don't know how to put this together.
The bottom line, however, is that this is all proof that the South is violent, backward and racist, and so are Republicans.
Why doesn't NPR care about its reputation for journalism?
Because they are in 'journalism' to "instruct," not to report.
No such thing as a discussion with virulent commies. I come from a long line of people that shot communists and othe Statists to keep other people free.
Why doesn't NPR care about its reputation for journalism?
Journalism is dead.
End of story.
If you are looking for neutral journalism, NPR and Totenberg are the wrong place to look. As if there's a right place.
One big problem today is that you cannot believe anyone about anything. Anything that comes out of the gov't or journalists is suspect.
People wanting to use race to reward one group and punish another group that had nothing to do with the racism of people 50 years ago: Not racist.
People wanting everybody to be treated equally under the law, all the time: racists.
Why doesn't NPR care about its reputation for journalism?
Sorry.. could not help laughing out loud when I read this line. The bias is so blatant that unless there is an uprising from all quarters about the state of the media in this country nothing good will come out of it. May be they are trying to intimidate Roberts and he seems easily intimidated judging by what he did with Obamacare.
NPR has really been Ground Zero for Leftist Media Bias since the 1970s and continues to this day.
Speaking of journalism, I just watched the pro-abortion "Catholics" on This Week discussing what the next Pope should do.
So, we can use the census data to apportion the number of Congressional districts and determine population counts of all the different ethnicities, but we can't use the same data to determine if we should stop VRA enforcement, because it is "inaccurate and out of date".
Yea, what a big pile of horseshit that argument is.
Yes its NPR; more important, its Nina Totenberg.
What an embarrassing display of eagerness to discredit Roberts!
You presume they are still capable of feeling embarrassment.
They aren't.
Like seemingly everyone in the political game, they are near to the point where they feel nothing but a damning mix of self-righteousness and anger.
Would it have changed Robert's arguments if he had qualified his statement saying for example "according to the census data of such and such year"?
The bottom line, however, is that this is all proof that the South is violent, backward and racist, and so are Republicans.
It's the risible "New Confederacy Theory"
They do not have to worry about being embarrassed because they are good and right, and you are a racist, bigot, homophobic, sexist, nationalist pig.
So what is the census good for?
The only question is will getting rid of the VRA help Republicans win more elections. That's a yes, so conservatives on the court will get rid of it.
She went out and talked to "Census officials" who told her that "these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states."
So census bureau is not doing its job?
She went out and talked to "Census officials" who told her that "these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states."
So Roberts did misconstrue the census results. He thought they might be accurate..
The only question is will keeping the flawed and inconsistent VRA help democrats win more elections. That's a yes, so the liberals on the court will vote to keep it.
One of the more interesting facets of the leftist strategy to portray conservatives and Republicans as, essentially, closeted supporters of the KKK, is that any instance of white gang violence must be agonized over eternally, but black or hispanic gang violence can always be explained away.
So, we must be constantly reminded of lynchings that happened 75 years ago, lest we forget the inherent evil of whites... but black and hispanic gang violence happening right now in Chicago, Watts and Newark... well, it's white's fault, or there's a good sociological explanation for it.
"...these numbers are simply not reliable for state-by-state comparisons.."
So let me get this straight. They have no fucking idea what the facts are, but they still want to keep this law in place.
It's difficult not to feel stupider, reading this. Do these people have negative IQs?
"Why doesn't NPR care about its reputation for journalism?"
Journalism isn't about mundane things like "facts" and "data". It's about "fairness" and "justice" (as perceived by the "Journalist"). At least, that's what the Washington Post thinks.
Tune in next week when Althouse reports that water is wet and the sky is blue!
Why is it anybody's business but the voter's, black or white, as to whether he will or will not turnout to vote? What is worst about a high ratio of white to black voters and best about a low ratio? Independent individuals are making, each, his individual uncoerced decision to participate or not participate in a voluntary, not a mandatory process. Isn't that the ideal?
So let me get this straight. They have no fucking idea what the facts are, but they still want to keep this law in place.
It's difficult not to feel stupider, reading this. Do these people have negative IQs?
Shut up. Southerners and Republicans are racists. Understand? If not, you're a racist.
Nina Totenberg swims out to meet troopships..
So data compiled within the last few years is unreliable (and yet is used by the government for any number of purposes), but the data from 40-50 years ago which is used to justify neverending Reconstruction intervention by partisans in the federal government is perfectly reasonable and not to be questioned.
ricpic said...
Why is it anybody's business but the voter's, black or white, as to whether he will or will not turnout to vote? What is worst about a high ratio of white to black voters and best about a low ratio? Independent individuals are making, each, his individual uncoerced decision to participate or not participate in a voluntary, not a mandatory process. Isn't that the ideal?
You're missing the point. If you ask voters to register more than 15 minutes before they vote, show ID, vote on election day, know where their own polling place is or know who the candidates are, then you are racist, bigot, homophobic, sexist, nationalist pig who must be monitored by the Feds.
She talked to someone. What's the quality of that exchange as well. Sound like water cooler gossip and NPR runs with that...
Erika, at the top, hits the nail on the head. Ann is simply whistling past the graveyard by posing such questions--get your head out of the sand, Ann..
I feel the same way about Nina Totenberg as I did about Daniel Schorr. Someday, something and it will probably be death, will free us from having to consider their biased ramblings hidden behind a wall of respected journalism.
FleetUSA said:
"Sound like water cooler gossip and NPR runs with that..."
Nanny Bloomberg and Yahoo's CEO Marissa Mayer said this week that a lot of important work gets done around the water cooler.
Another entry in the Annals of Questions That Answer Themselves.
Yes, as Andrew Klavan (and others) have pointed out, most leftists arguments boil down to "Because... shut up!"
See, Ann attempts to maintain her "cruelly neutral" "non-partisan" academic stance at all costs to actual reality. I much prefer to march under Churchill's banner: "I will not be neutral as between the fire and the fire brigade."
Journalists lie. Judges don't really judge fairly. Statistics are unreliable at best and can mean whatever their user decides they mean.
Nothing new here.
We all get it. We know exactly how we're being manipulated. The manipulators don't even bother to hide their tracks or deny their lies.
This is new and very disturbing.
So, we must be constantly reminded of lynchings that happened 75 years ago, lest we forget the inherent evil of whites... but black and hispanic gang violence happening right now in Chicago, Watts and Newark... well, it's white's fault, or there's a good sociological explanation for it.
Worse, they just ignore the very real race war being conducted by the MS-13 affiliated Mexican gangs against blacks, driving them out of neighborhoods with assassinations and thuggery.
But, we need to only talk about Devil Whitey, who doesn't know any racist gang-bangers and has never oppressed anybody or wasn't even alive in the 1950/60s.
The judges don't know stats so Roberts' criticisms are probably poorly grounded.
Didn't the SCOTUS reject the push to use statistics in conducting the census? I can see why they might have had constitutional concerns, but as a result, we now have less accurate census data every ten years. Way to go SCOTUS!
I don't see why, when it comes to narrowly, technically "correct" fuck-ups like this, SCOTUS doesn't just recommend a remedy back to Congress and have them initiate an amendment. It's pretty pathetic that we're restricted to using 18th-century mathematical conventions to work out a 21st-century accounting procedure.
AT EASE, Sgt Ted, you're getting dangerously close to reality-based too-close-for-comfort-for-lefties-and-Obama seditious talk. Big Sis has big ears...know your place..
First, those lynchings were more like 100 years ago.
Remember, they didn't have TV, radio, or movies back then.
Second, somebody actually believes Nina Totenberg has a rep for accuracy?
Totenberg and NPR want Obama to succeed at all costs, simple as that. They are NOT journalists and no longer care about their reputations.
There goes an appellate judge asking questions from the appellate record, conclusion Racist!
To quibble:
Abstruse vs. obtuse
That is the question.
"Why doesn't NPR care about its reputation for journalism? "
One reason could be that they don't have a reputation for journalism. Another could be that in the current debased state of the profession, they think what they're doing is journalism and they're very good at it. The best and the brightest.
Ritmo you ignorant slut (that's an SNL line not an attack so don't get huffy). Read the article. "Roberts comments and conclusion appear to be taken from a census SURVEY...". (emphasis added). This IS the statistical approach of which you approve, NOT the 18th century full count you deride.
Why is so little attention being paid to the behavior/questioning of Sotomayor and her side kick (Kagan)?
http://www.breitbart.com/Big-Government/2013/03/02/Sotomayor-Kagan-Try-to-Browbeat-Scalia-over-Voting-Rights-Act-Argument
I understand this account was taken from a certain perspective, but I truly do not see why it is being ignored, especially here.
It's not too late to get your "Nina Totin Bag," a tchotchke offered by NPR to those who give it gifts.
The Wiki> profile of her is replete with fairly wild examples of her close connections to the liberal judges and the Democratic Party.
The comments on the NPR page with the article are funny. The Chief Justice's attack, apparently made without vetting the statistics, or understanding them, or ignoring their flaws, was not a quest for truth. Rather, it was an attempt to justify the conclusion he's already reached. As a former law clerk to an appellate court judge, I don't find much to admire in the Chief. As if Totenberg doesn't write articles to justify a conclusion she's already reached.
Why doesn't NPR care about its reputation for journalism?
Well, if the New York Times and the Washington Post don't care about their reputation for journalism, why should NPR?
Totenberg has been one of NPR's leading political hacks for decades. Perhaps without her the tone of the organization would have been measurably closer to news-reporting than to liberal cheerleading. Still, it's valuable work to point out, as Professor Althouse has done, some of Totenberg's hackery in action.
Read Totenberg's bio, and had forgotten how long ago she had been widowed from one of our less illustrious Colorado politicians, Floyd Haskell. One of those who decided not to move back here after losing reelection, but rather, to stay in D.C. and peddle influence. Oh, and Obama appointed her sister to a federal judgeship.
What must be remembered about N(L)PR is that it is primarily funded by contributions, mostly from their beg-a-thons, and their contributor base is well meaning affluent liberals. They get their jazz and symphony music, their high brow left leaning commentary, Garrison Keiler, etc., and in trade, they give a little, and feel good about themselves contributing to American culture.
The amazing thing to me is that anyone actually thinks that they are anything but left leaning hacks and, in her case, well connected liberal political operatives.
NPR: Knowitall Radio in action:
Because they are in 'journalism' to "instruct," not to report.
Bingo!
The strange new respect Roberts earned with his tortured logic in the Obamacare case didn't last very long.
It's hard to care about your reputation for journalism when you're busy carrying buckets of water for the administration.
i.e., "When you're up to your ass in alligators, it's hard to remember the original objective was to drain the swamp."
Why is the Solicitor General referred to as "General" in these transcripts? It's ridiculous to refer to someone's title by the adjectival part of it alone.
It's like referring to the junior senator from a state as "Junior".
The Professor frequently cites issues with headlines; they are confusing, grammatically inept, agenda driven and often, as in this case, not truly representative of the story it serves.
Being ignorant of the processes of news media (except for the assurance that they have many layers of fact checkers) my question is who writes the headlines? I take it that the reporters don't. Is there a job of headline writer? At NPR, is it done by journalism interns or low level newbies? I'd like to know where the breakdown is.
Chip S:
I believe the same is done for the Surgeon General and Attorney General. I have a friend who was career FBI and was a Special Agent in Charge for two FBI offices in his career. I remember him telling me stories in which he addressed Janet Reno as "General".
Of course when addressing Eric Holder you also are expected to first bow and request permission to kiss his derreiere.
I salute General Mills with every bowl of Honey Nut Cheerios.
"Why doesn't NPR care about its reputation for journalism?"
It does care about its reputation for "Journalism." What it doesn't care about is its reputation for "reporting."
Totenberg was so busy straining at Roberts' gnats that she apparently swallowed a camel from Kagan:
If Congress were to write a formula that looked to the number of successful Section 2 suits per million residents, Alabama would be the number one State on the list.
All else being equal, the number of successful Section 2 suits is going to be proportional to the number of black voters. Is the VRA supposed to target states for discriminating against black voters-- or merely for having them?
Garage said: The only question is will getting rid of the VRA help Republicans win more elections.
No. The only question is will getting rid of the VRA help Republicans win more elections fairly.
I'm trying to find the actual numbers, CWJ (and no offense taken at the funny SNL remark), but can't find them. I'm finding news on the rhetoric from the MA SOS, but can't see what they're basing their opinions on.
My guess is that MA might have greater disparity between a registration rate that is high for both whites and blacks (say, in the 90s) whereas MS might have less disparity in registration rates that are still nonetheless both low. But again, without seeing the numbers, who knows?
Chip: Roberts' decision in Obamacare was correct. Machos agreed. It's not his job to opine on whether the legislation looks good to him, but on whether any provision of the Constitution allowed it.
I'll try to be open-minded to the way this should be decided, but Jim Clyburn's admonition to Scalia caught my notice:
"When you have in 2012 ... states making changes to their laws that you can look on their face and see that these changes will make it harder for minorities to have their votes affect the results that they intend, you say that we don't need [the Voting Rights Act] anymore? Is this some kind of entitlement?" Clyburn asked. "Well, the Constitution of the U.S. is an entitlement for everybody."
I hope they're careful with the way this is decided.
Ms. Totenberg's argument is apparently that Mississippi cannot be proven better than Massachusetts because of the large standard errors. Perhaps so. But given that MS ranks better than MA, it is reasonable to conclude that MS is at least as good as MA. So why should MS have burdens that MA does not?
garage mahal said...
"The only question is will getting rid of the VRA help Republicans win more elections. That's a yes, so conservatives on the court will get rid of it."
See what you did there? You equated Republicans with "getting rid of the VRA [to] help[ing] Republicans win elections..."
Nobody is talking about getting rid of the Voting Rights Act, garage. Section 2 of the Act, where all of the prohibitions against voting discrimination reside, is unchallenged. Only Section 5, which requires the goofy "preclearance" routines on the part of some areas of the country, is at issue.
But when low-grade Republican-hating idiots like you, garage, keep repeating the mantra that Republicans and/or conservative judges want to get rid of the VRA, you are fulfilling what the partisan left really wants. To equate Republicans with the kind of racism that Democrats engaged in decades ago.
Republicans voted overwhelmingly and sometimes unanimously to reauthorize the Voting Rights Act. Because they wanted to make sure that morons like you, garage, couldn't credibly say that Republicans opposed something that was generically thought of as "voting rights."
In fact, Republicans hated Section 5 all along. But since it was part of the Act, and the Act was one that Republicans felt they could not vote against for the reason I've just described, the Republicans held their noses and voted for it.
But you are again talking about the Republicans, and conservative judges, opposing the Voting Rights Act. You're wrong, garage, and you're stupid.
The "conservatives on the court" won't be getting rid of the Voting Rights Act. You and the others on the left ought to be slapped down for making the suggestion. You are the very embodiment of the reason that so much of America does indeed regard the Voting Rights Act as a racial entitlement. Because you think of any alteration of any part of the law as an erasure of voting rights, when nothing of the kind is true.
So why should MS have burdens that MA does not?
Apparently Nina T. is a Bayesian, w/strong priors.
It's interesting that NPR and Totenberg perceive disenfranchisement, but do not see it when votes are cast by illegal aliens and other ineligible people, or are manufactured whole cloth behind the scenes.
Actually, they have demonstrated a very selective concern for civil rights. It started with their local domains, and local corruption, but in the last five years or so has become a prominent feature at the national level.
Well, they wanted to establish credibility for their long-running narrative, and they were willing to disenfranchise and corrupt individuals, institutions, and the government to achieve it.
More Black Men in Jail Than in College? Wrong
And so another myth was born, to be exploited for political, economic, and social opportunity. But, what are the side-effects? These are presumably tactical measures, but with long-term consequences.
O Ritmo Segundo,
Didn't the SCOTUS reject the push to use statistics in conducting the census? I can see why they might have had constitutional concerns, but as a result, we now have less accurate census data every ten years. Way to go SCOTUS!
It did so because the plain text of the Constitution calls for an "actual enumeration." Not a guesstimation, but actually going out and counting the whole population. Yep, I'd call that a "constitutional concern." I can't think of many things clearer in the document, apart from the minimum ages for Congress and the Presidency and the terms for House and Senate.
Shouting Thomas - "So, we must be constantly reminded of lynchings that happened 75 years ago, lest we forget the inherent evil of whites... but black and hispanic gang violence happening right now in Chicago, Watts and Newark... well, it's white's fault, or there's a good sociological explanation for it"
=================
Interestingly, a sizable number of those lynched were white (30%), and a sizable number of blacks lynched were by angry black mobs exacting "street justice" for black on black crimes.
The truth is that you can't simply term that non-state sanctioned way of achieving "swift and sure justice" - as a white on black for the purpose of racial intimidation.
It's more complicated.
Many blacks were ardent supporters of lynching, inflamed by the same passions as whites.
The anti-lynching campaign was actually launched by the leftist Jews in charge of the NAACP in the early years 1911-1930. (Jews were in charge up until the mid-60s) To them, it was too similar to the pograms that Jews had faced in Russia and Europe when the native's resentment of Jews periodically boiled over.
The Jews of the NAACP wanted all matters settled by lawyers in courts.
Part of that was outreach to "educate" their black children, as they saw them.....that blacks had to stop stringing up their own thugs...and cast lynching as a vehicle by which evil whites oppressed them like the Czar repressed the Jews.
And use media to cast lynching as purely about evil whites killing innocent blacks "Down South" and in the ignorant white Christian Midwest. (Most studies show that there were definitely innocents lynched, that it was used as racial imtimidation....yet the bulk of the people lynched by blacks or whites were guilty of what they were being hanged or shot for.).
Shouting Thomas is right that part of what is going on in Chicago and other black communities is that blacks are killing other blacks in modern lynchings because there is little belief that the lawyers and courts outside the 'hoods will get real justice done. Same as when blacks spitted and roasted a black nere' do well caught after robbin' and rapin' and killing some of their womenfolk in 1915 Alabama.
Of course, like lynching, the problem of black on black and black on white violence has other drivers - black thugs reenacting Hatfields and McCoys, violence to rob or do rapes of opportunity, primitive impulsiveness unchecked by liberals and their beloved lawyers and courts. (Awwww, the poor troubled youth acted out and shot a few people!! He needs more education and Obamabucks..)
Why doesn't NPR care about its reputation for journalism?
NPR only has a "reputation for journalism" among left-wingers, and left-wingers aren't going to question it on this point.
He didn't. Totenberg is doing cleanup work. She went out and talked to "Census officials" who told her that "these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states."
If the numbers are such junk, why are we paying for them in the first place? I'm kind of appalled.
So two of the original agencies of the US Government, the postal service and the census bureau, are miserable failures. The head of the census bureau should be embarrassed to have to admit that after 200+ years, we still can't count people reliably.
But hey kids! Not to worry! You get to see your own miserable failure kickoff! Yes, the people who made “unintended consequences” a reality for so many welfare families, head start children, Medicare recipients, and so many more programs that can never be cut or really called into account, now bring you obamacare!
The statement that "the margins of error are too high" to infer that meaningful differences exist is equivalent to the statement that "the differences b/w states are too small" to make that inference.
It's not a matter of "junk numbers". It's a matter of determining how fine are the distinctions we want to be able to draw in choosing the sampling rate.
The head of the census bureau should be embarrassed to have to admit that after 200+ years, we still can't count people reliably.
The voter-registration data have absolutely nothing to do w/ the decennial census, other than that both are collected by the Census Bureau.
The data that Nina T. is referring to come from the Current Population Survey, which is a recurring survey of a small random sample of the population. It is completely different from the decennial head count mandated by the Constitution.
Just to elaborate slightly on Chip S' last two comments, its not just sample size but also sample design. The CPS sample is stratified by state; meaning it is intended to allow state by state estimates.
I was going to post that without state by state stratification, the ability to reliably estimate individual state statistics would be compromised regardless of sample size. But since that's not the case, I find the comments concerning misuse of the statistics somewhat confusing. The CPS consists of 50,000 respondents; an average of 1,000 per state. Mass is a fairly populous state. If the sample is in all other ways good, and you can't find meaningful differences with a sample size of 1,000, then the differences in all probability aren't particularly pronounced.
Couldn't we have Nate Silver mutter an incantation over the census numbers so the Reality-Based Community will accept them?
If the census data is unreliable or useless, shouldn't they imporove or drop colating it?
Totenberg is frequently a silly twit.
are you talking about Nina gee-i-apologize-for-using-the-phrase-'Merry Christmas Totenberg (on The McLaughlin Group) ?
THAT Nina Totenberg?
Bruce,
"The amazing thing to me is that anyone actually thinks that they are anything but left leaning hacks"
Ummm, I think you left out the (NTTAWWT)?
Cedarford:
Interesting. I know that most of the violence suffered by Blacks is at the hands of other Blacks. I did not, however, consider that it may follow from a self-policing action. They have problems within their largely insular communities. They don't trust outsiders to treat them fairly. So, they have effectively established "sovereign" domains where they govern. Unfortunately, this isolates them from the larger economy, and it alienates them from the larger population. Their self-destruction was assured.
Unfortunately, the people who are selected to represent them in the world at large, are more interested in exploiting the situation for their benefit, than to address the causes and facilitate their integration with the rest of the population. It must suck to be classified a minority interest, but not receive the benefits which accrue with that designation. I would imagine people develop an unhealthy cynicism, which acts as a positive feedback to propagate their condition.
It did so because the plain text of the Constitution calls for an "actual enumeration." Not a guesstimation, but actually going out and counting the whole population.
Referring to statistics as "guesstimation" does not reflect well on your education.
Re: Chip S:
The statement that "the margins of error are too high" to infer that meaningful differences exist is equivalent to the statement that "the differences b/w states are too small" to make that inference.
If they went with the standard explanation that there's no statistical difference between the two populations, that would have been fine. These officials have specifically called out that their numbers have such "high margins of error" that they can't be used for between state comparisons. If they have such high margin of error, why bother collecting them in the first place? And if we are going to collect them, they ought to do a better job of picking a sample that's large enough that they don't have such "high" margins of error.
Balfegor, Totenberg is reporting her account of what she was told. It is not a direct quote.
Here is what she said:
Break that up into roughly proportional samples in each state, Census officials say, and it is really not possible to compare states because those with relatively low minority populations have a much higher margin of error.
As it's written, this makes no sense. Therefore, my presumption is that it's more likely a reflection of Nina T.'s ignorance of statistics and political bias than an accurate quote from any "Census officials", unless those "officials" are themselves partisan appointees.
It's always possible to compare the means of two samples of different sizes. This is a trivial statistical test that is taught in every Intro to Stats course in existence.
If the differences b/w any two sampled populations are large enough, they're statistically significant, for the simple reason that "significance" is based on the relevant ratios of sample means and sample standard errors.
What a fair-minded person who understood statistical inference could say in this case is that there is no statistical basis for inferring differences in minority-voter turnout b/w MA and MS.
Roberts' observation is a perfectly fine way of making this point, b/c the diff b/w MA and MS is the opposite of what you'd expect if MS was doing worse than MA. So the only logical possibilities based on the two samples are (1) MA is in fact worse than MS, or (2) there's no statistically significant diff. b/w them. Either way, there's no evidence that MS is somehow worse than MA, which is what really matters in this case.
Now If the observed diff. is so big that it would be important if it were statistically significant, then it's worth the cost of bigger samples. But mere statistical insignificance itself doesn't tell us that.
Nina Totenberg is...unreliable. I believe she was the source of Anita Hill. (I could be wrong on that.)
The Solicitor General is the general solicitor for the nation, not the general of solicitors. DAMN YOU, TRANSCRIPTS!
Re: Chip S:
As it's written, this makes no sense. Therefore, my presumption is that it's more likely a reflection of Nina T.'s ignorance of statistics and political bias than an accurate quote from any "Census officials", unless those "officials" are themselves partisan appointees.
Fair enough. I am always sympathetic to explanations that involve journalists' ignorance.
Post a Comment