John Roberts’ Curious Voting-Statistics Sophism Misconstrues The Census Report’s Statistics by Failing to Consider Key Statistical Deviation Facts and Fails To Consider WHY Massachusetts Blacks Might Be Voting In Lower Percentages Than Mississippi Blacks Are, Even IF They Are. [UPDATED]

In a blog post titled “In Voting Rights Arguments, Chief Justice Misconstrued Census Data” on NPR’s website, veteran NPR Supreme Court correspondent Nina Totenberg deconstructs a sophism offered by John Roberts at the oral argument on Wednesday on the continued constitutionality of a key section of the Voting Rights Act of 1965, which Congress has extended several times, the last time, overwhelmingly, in 2006.  Totenberg writes:

At the voting rights argument in the Supreme Court on Wednesday, Chief Justice John Roberts tore into Solicitor General Donald Verrilli, grilling him on his knowledge of voting statistics.
The point the chief justice was trying to make was that Massachusetts, which is not covered by the preclearance section of the Voting Rights Act, has a far worse record in black voter registration and turnout than Mississippi, which is covered by Section 5 of the act.

But a close look at census statistics indicates the chief justice was wrong, or at least that he did not look at the totality of the numbers.

Totenberg goes on to say that the statistics Roberts used were taken from a lower-court dissenting opinion, statistics that in turn were taken from a 2010 census report.  “But,” Totenberg says, “upon close examination, the numbers are less than reliable, according to the Census Bureau itself.”  She explains:

Here’s the deal. The Census Bureau does voting surveys to look at voting patterns nationwide, but the survey is based on a very small sample. Most recently, in 2010, the survey looked at 94,208 voters nationwide. 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.

The number of black citizens eligible to vote in Massachusetts is 236,000, while it is 721,000 in Mississippi, more than three times that number. Therefore, according to Census officials, when looking at the estimated turnout rate in Massachusetts, the voting percentage for African-Americans at first blush is estimated at 39.3 percent. But the margin of error is 11.5 percentage points, meaning that the black voter turnout actually could be as high as 50.8 percent (or, conversely, as low as 27.8 percent). 

Now, look at Mississippi, where black turnout is listed at 48.7 percent. But because of the large size of the African-American population that was sampled, the margin of error is only 5.4 percentage points.

That means that factoring in the margin of error, the black turnout rate in Mississippi could be as high as 54.1 percent, or as low as 43.3 percent.

So, if you factor in the margins of error at their extremes — with Mississippi at the low end and Massachusetts at the high end — Mississippi could have had a black voter turnout rate that was 7.5 percentage points lower than Massachusetts.

Bottom line, as Census officials told me, these numbers are simply not reliable for state-by-state comparisons because of the high margins of error in some states.

Yep, John Roberts, it turns out, is no statistician.  But neither is he, well, a social scientist. Or even a moderately sharp observer of politics.  At least, not of Massachusetts politics. Where, y’know, Democrats usually win with lopsided victories in, say, House races in predominantly black Congressional and state-legislative districts; where the outcome of the state’s electoral college presence in presidential elections has not been in doubt for decades; and where neither Ted Kennedy’s nor John Kerry’s reelection was even remotely in doubt since, I guess, the 1994 Ted Kennedy/Mitt Romney campaign. Totenberg’s report makes clear that voting in the hotly-contested statewide race to replace Kennedy, who died in office–a special election in late 2009–was not part of the census report, which measures voting only in national elections.  

So, might not the question of what’s at stake in a particular election matter, even when you’re a Supreme Court justice who’s spent decades wanting to see the Voting Rights Act’s demise and will not be fussy about the grounds you state for your decision?  I mean, just to make it look sorta rational?

Most of the media attention concerning that oral argument focused on what were truly shocking comments by Scalia in which he said both that voting rights are a racial entitlement and, equally stunningly, although gaining less attention, that the motive of members of Congress in voting to “perpetuate” that “racial entitlement” in extending the Voting Rights Act–political reasons, he said–is appropriate reason for the Supreme Court to refuse to give the usual “deference” (here, legalese for benefit of the doubt) to Congress’s legislative decision after weighing extensive evidence, even if the motive was itself not unconstitutional.  

This latter–that the motive of members of Congress, which in this case no one claims was an unconstitutional motive (e.g., a racist motive or a religious-discrimination motive)–purports to give the Court the constitutional authority to reject Congress’s legislative choices–is so obviously bizarre and dangerous that, in this case, it likely won’t garner agreement by any other justice. But that oral argument was the second time in less than a year that Scalia offered this argument, and the first time he did so, he was joined by three other justices.

The case in which this occurred was the ACA (Obamacare) case.  In his dissenting opinion, he claimed that because a majority of the Court was voting to strike down as unconstitutional one section of the statute–the section concerning the consequences to states of refusing to agree to accept the expansion of Medicaid–the Court must strike down the rest of the statute, not because a majority thought the rest of the statute also was unconstitutional (a majority did not), but because a few members of Congress who voted for the statute might not have voted for the statute if that Medicaid-expansion part was not part of it.  The ACA passed the Senate with no votes to spare, see, so, well, I mean, who’s to say that without the Medicaid-expansion part, the statute would have passed at all?!

Rest assured, though, that this, like sooooo many other movement-conservative gimmicks, would be entirely discretionary with the justices, or lower-court judges, case by case.  A budget bill that conservatives favor, and that’s enacted in a close vote, would not be strikeable in its entirely even if it contained some part that a majority of the Court thinks is unconstitutional.  But a budget bill that disfavored by conservatives, and enacted in a close vote, might be strikeable in its entirety if one part is deemed unconstitutional, since ya never know whether it was that log-rolling of the sort that Scalia, Kennedy, Alito and Thomas objected to in the ACA, and that involved the unconstitutional provision, that enabled the passage of the budget bill.

This, by the way, apparently was the bridge-too-far that began Roberts on the road to a change of heart, and change of vote, on the constitutionality of the individual-mandate provision. Reportedly, it really offended and scared him.  But it got the votes of four of the nine justices.

This time around, though–in the Voting Rights Act case–Scalia’s apparently trying only to use this motive-matters argument to strike down a key section of the Act, not the whole thing. Which makes sense, since log-rolling wasn’t the reason for the 2006 reenactment of the law, and, as I said, the law was reenacted by overwhelming votes in each house; 98-0 in the Senate. But luckily for Scalia and friends, among them Roberts this time, there is that census report.  

And Nate Silver hasn’t been asked to analyze it. And he probably won’t be.

UPDATE: OH. WOW.  Politico reported last night, in an article titled Massachusetts official slams chief justice’s comments on Voting Rights Act:

The problem, Massachusetts Secretary of State William Galvin says, is that the data does not back up Roberts’s claim.

“It’s just disturbing that the chief justice of the United States would spew this kind of misinformation,” Galvin told POLITICO.

Galvin’s office assumes that Roberts was going off U.S. Census Bureau data, which is one of the only national datasets on voter turnout by race, but they say the 2010 numbers don’t support what Roberts is saying.

“He’s wrong, and in fact what’s truly disturbing is not just the doctrinaire way he presented by the assertion, but when we went searching for an data that could substantiate what he was saying, the only thing we could find was a census survey pulled from 2010 … which speaks of noncitizen blacks,” Galvin said. “We have an immigrant population of black folks and many other folks. Mississippi has no noncitizen blacks, so to reach his conclusion, you have to rely on clearly flawed information.”

The 2010 tables show that Massachusetts does have a high discrepancy between turnout of white and black voters, but is in line with several other states, including Minnesota, Kansas and Washington, which actually has a wider ratio. The states are also similar on registration numbers. Additionally, the margin of error on each of these states’ data is over 10 percentage points, and many states on the list had populations of blacks so small, data wasn’t even available.
“We reached out to academics at many institutions … and they could find no record either, they were puzzled by [Roberts’s] reference,” Galvin said.

So Roberts was gamed by an incompetent or intellectually dishonest lower-court judge, from whose dissenting opinion Roberts took the bait, the hook, the line–and the sinker. It’s fairly commonplace for lower-court judges to misstate or omit key evidence.  Sometimes, it’s because the law clerk they’ve assigned to do their work for them doesn’t bother to actually nail down or verify facts put into briefs, and sometimes it’s just that they know how they want to rule, and just cherry-pick truncated statements of fact to support their chosen conclusion.  And sometimes–and believe me; this is true–they just fabricate a fact out of nowhere.

But, really–the chief justice of the U.S. Supreme Court using a statistic on voting that includes non-citizens as eligible to vote, because he just lifted a surprising statistic from a lower-court judge’s dissenting opinion, without first looking into its plausibility, much less its actual accuracy?

What’s next? The chief justice relying unquestioningly on a Bob Woodward report?