Image and video hosting by TinyPic

Thursday, June 16, 2005

Manure from Manjoo: A response

In the "comments" section, Salon writer Farhad Manjoo responded to yesterday's call for a Salon boycott, which I issued in the wake of his piece attacking vote fraud investigators. In the interest of fairness, I'll repeat his words in full here. (Italics indicate when he has quoted me.)

1) Actually, Manjoo replaces the "rBr" nomenclature with a new label: "exuberant Democratic responder."

That's a lie. I never say those words. You just made that up. And it's pretty astonishing that you do so in the context of questioning my ethics.

2) Are any Republicans of your acquaintance shrinking violets? Turn on your radio: Do the voices of reaction sound even slightly embarrassed or cautious? Democrats are the ones who have been cowed into silence by the violent brownshirts of the right.

Saying it's horseshit doesn't refute the studies done by pollsters showing this to be the case (studies cited in my article). Can you offer any response to those studies? And can you explain why exit polls -- not only in 2004 -- consistently show a Democratic tilt? You should be able to support your calling it horsehit. You do none of that here. (And if your response is that you can't believe the pollsters' studies because they're corrupt, tell me why you believe their polls.)

3) And why were exit polls accurate in this country until the ascension of the Bush dynasty? Why are they still considered extremely accurate in Europe?

That's wrong. You're making that up. Exit polls showed Clinton winning bigger than he did in 1992 and 1996. The only reason people didn't cry about it was that those races weren't close, and Clinton still won. As Mark Blumenthal points out, you can see proof of this in the movie "War Room." See how they figured based on the exits Clinton would win a landslide. Of course, he did not.

Don't lie to your readers about the history of exit polling, Joseph. 2004 was not the first year the exits were off.

4) If you look carefully at the work of Mitofsky and Liddle, you will see that their theories accept as a given the proposition that the count was accurate. How can we accept an explanatory scenario from someone who stipulates the very point under dispute?

Again, you're wrong. You're either lying or mistaken. Liddle does not accept this as a given, Joseph. You just made that up.

5)According to Manjoo, anyone who dares to raise these common-sense issues deserves to be damned as a wild-eyed lefty. In the past, Salon itself has been on the receiving end of such derision. How dare they lob similar accusations at those who do the progressive work Salon now prefers to ignore?

That's silly, Joseph -- I never suggest that. My article says only this: The evidence showing that the exit polls were off is pretty good. I say nowhere that Republicans are incapable of fraud; that raising the issue of fraud is crazy; or that we shouldn't reform the election system.

In fact the reason to refute the refutable theories of fraud is precisely to push forward reform. Salon has been pushing for election reform since the 2002 election. Search for "salon diebold" in Google. You'll see were were one of the first major outlets to write about that company and the dangers of electronic voting machines. I'm even one of the good guys in Bev Harris's book, for Pete's sake.

But if you want real reform you need to do it in a bipartisan way; you're not going to get any moderates on board by yelling, without evidence, that the last election was stolen. Which is exactly what you're doing in this post (with, as I've shown, not even the facts to back you up).
Where to start?

Let's begin with Bev Harris -- "Saint Bev," as some call her -- cited by Manjoo as though I should be very, very impressed with her. As indeed I once was. Then she announced that she had found discarded poll tapes in Florida which would prove that fraud had been committed. Quoth Bev: "The pattern was very clear. The anomalies favored George W. Bush. Every single time."

In late November, the Sainted One promised a "major announcement" concerning those tapes. We're still waiting for it. Whenever the topic comes up, she retreats into a Trappist silence -- and she even refused to show the tapes to MSNBC's Keith Olbermann.

Since she refuses to drop the other shoe, forgive me for doubting that she ever had a shoe to begin with. Sorry, Farhad, but I'm no longer impressed by that woman -- just confused.

Manjoo tries to convince his readers that "fraud freaks" have concentrated on exit poll controversies to the exclusion of concerns he considers more vital. That accusation hardly applies to me, to Brad Friedman, to the many Democratic Underground posters, to the writers at Raw Story, to the heroic workers who oversaw the Ohio recount, and to any number of people who have devoted time and energy to the discussion of all aspects of vote fraud.

You ask me to fire up Google, Farhad? Fine.

Let's type in "Manjoo" and "Clint Curtis." Let's type in "Manjoo" and "John Conyers" and "Ohio." Or "Manjoo" and "Katrina Sumner." Or "Manjoo" and "Richard Hayes Phillips." Or "Manjoo" and "Snohomish." Or "Manjoo" and "Georgia10." (Whoever she is, she wrote an invaluable piece on the shennanigans in Ohio.) Or "Manjoo" and "Triad." Or "Manjoo" and "Ahmanson." Or "Manjoo" and "recount" and "Ohio." Or "Manjoo" and "Carter" and "Baker." Or "Manjoo" and "machine" and "switched" and "vote." Or...

Well, why go on? This blog and others have discussed many, many aspects of the vote fraud controversy which Manjoo has left unaddressed. True, the combination of "Manjoo" and "African-American" and "disenfranchise" will bring up one relevant piece by Manjoo -- but it was written before the election.

In fact, Manjoo is the one who has concentrated on exit polls. This isn't a case of the pot calling the kettle black. More like the tomato calling the cucumber a red.

His pathetic cries of "liar liar" are similarly disingenuous.

In my reading, Mitfosky and Liddle have provided no evidence that either Democratic responders were enthusiastic or Bush Responders were reluctant. They have merely indicated that such might be the case -- leaving it for the propagandists to shift "might be" to "is," as Manjoo has done. For all the abstruse arguments over statistical methodology, the argument comes down to this: If we posit that Bush responders were under-represented, then the exit poll disparity is explained. Again, this scenario demands that we stipulate as a given the very point we are trying to determine.

Manjoo presents no real proof that Kerry voters were over-represented. But I have often discussed proof (I'll go so far as to say hard proof -- at least, I've seen no persuasive refutation) that, in fact, the exit polls over-represented Bush supporters. Not only did Kerry win, he won by a larger margin than most would consider possible.

Regular readers will forgive me for repeating one simple, damning fact again, but as the old song says -- if the man don't listen, you got to yell a little louder. Now, I'm fearful that Manjoo might lapse into a "Deiter" impersonation ("Your qvestions haff become tiresome; now is ze time on Shprockets ven ve dahnce!), which is what most people do the moment they are asked to respond to this point, which I consider vital. So let's use boldface letters. That way, even Manjoo can't miss the message:

In 2004, the exit pollsters also asked the voters about their choice in 2000. The majority of respondents said that they had pulled the lever for "Bush" in that election. Yet in 2000, AL GORE WON THE POPULAR VOTE.

Got it?

If you like, I can use larger type. Italics. Maybe upload a sound file. Because I've been making this point for months now. So has "Truth is All" on DU (at great length, and in boggling detail). So have others.

Yet Manjoo and his fellow ostriches refuse to address the issue!

He refuses to explain how, in the known political universe, the exit pollsters might encounter so many people resistant to Bush's charms in 2004 who nevertheless brag about (or confess to) voting for Bush in 2000. If Bush voters are by nature reluctant to speak to those eeee-vil pollsters, why doesn't this reluctance color the question about Bush/Gore?

Elizabeth Liddle, I can happily report, did make a stab at responding (in these pages) to the 2000/2004 problem. I respect her for not pulling a "Deiter." But as my readers know, in this instance, Febble was feeble, offering hazy guesswork where fact was needed.

Good Sir William of Occam remains our most reliable guide: The simplest explanation to cover all the facts is likeliest to be true. In this case, the simplest explanation is that the exits pollsters over-represented the Bush vote, not the Kerry vote. The evidence favoring this idea is a lot stronger than the rBr (or eDr) nonsense peddled by Mitofsky and Manjoo.

Incidentally, Manjoo calls me a liar for seconding the opinion at DU that his argument can be summarized by the sobriquet "enthusiastic Democratic responder." (I've always preferred the term "chatty Dem theory," which almost caught on...!.) But his own cyber-ink proves beyond rational debate that he is, in fact, pushing what we may call the "eDr" scenario. (Granted, in yesterday's post I should have said "Manjoo would have us replace..." instead of "Manjoo replaces..." I do apologise for writing too rapidly. But the basic point still stands.)

Regarding previous exit polls:

Exit polls showed Clinton winning bigger than he did in 1992 and 1996. The only reason people didn't cry about it was that those races weren't close, and Clinton still won. As Mark Blumenthal points out, you can see proof of this in the movie "War Room." See how they figured based on the exits Clinton would win a landslide. Of course, he did not.
In this instance, I was careful in my wording. I said that exits were considered accurate until the ascent of the Bush dynasty. Diebold and ESS have been up to their dirty numbers for a while now. Thank you for helping to buttress my point!

Saying it's horseshit doesn't refute the studies done by pollsters showing this to be the case (studies cited in my article). Can you offer any response to those studies? And can you explain why exit polls -- not only in 2004 -- consistently show a Democratic tilt?
"Cited"? Maybe I'm missing something, but I've been clicking the links in Manjoo's June 15 piece and have yet to find any citation to an exit poll study that goes back before the rise of the Bush dynasty. Yes, I have seen studies which claim to demonstrate a consistent Democratic tilt in the exit polling since 1988. But these studies operate under the presumption that the "actuals" are irrefutable and the exits must therefore be mistaken. Studies that stipulate the very point under question are of little value.

The study Manjoo does cite discusses the display of folders and pens bearing network logos in an experimental situation. This experiment is of no real value to us. As Manjoo himself notes, pollsters in 2004 did not flash logos. Since no evidence indicates any voter reluctance to speak to non-logo-bearing pollsters, his point disappears into the vapor.

And Manjoo still doesn't respond to the question about the accuracy of exit polls in Europe, but not here.

Neither does he mention many other pertinent points, such as the fact that vote tabulation companies have a disturbing history of bribing election officials and offering them comfy sinecures, in order to get their hackable machines into position. Or this administration's efforts to impede international observation of the 2004 vote. Or Leto's study of Snohomish county, where the electronic vote differed substantially from the paper vote -- a difference inexplicable by any theory other than vote fraud.

Neither does Manjoo offer any suggestions as to how we might double-check our highly-questionable paper-free compu-vote. If we blithely toss out any exit polls that conflict with the official story, what else do we have?

(By the way, I'm not the only one to detect the distinct odor of rodent when Manjoo walks into the room. Readers interested in Bob Somerby's take on Manjoo should click here.)

About the larger question of Salon's current value to progressives:

I'm afraid the famed cyber-zine is at a familiar crossroads. Here in Los Angeles, the free (and formerly feisty) L.A. Weekly reached a similar point in the late 1980s. That's when a publication which had done so much to stop Reagan's planned war in Central America presented us with an endless parade of inane cover stories about Madonna's cultural impact or the sexual confessions of that strange lady who later married Elmore Leonard.

Salon, which was once the place to go for those who wanted the truth about the Right's war on Clinton, now prints very little cutting-edge investigative reporting. Lately, I've gone there mostly for the movie reviews. But in the Bush economy, I can't really afford movies. So why read the reviews?

Finally: I don't want this blog to devolve into a pissing match between one Salon writer and myself. Manjoo delivered a cup and I have responded with a quart. If the urine overflows the gallon mark, I'll restrict further emissions to email.
Joseph: There seems to be a great deal of this going on lately -- would-be Right-wing "pundits" who lack the facts or investigative background to even comment meaningfully on their chosen topic nevertheless getting all "lathered up" at the notion of any mere mortal calling them to task. One can only wonder if this isn't yet another case of "Your Tax Dollars at Work [Against You]", courtesy of the Bush Machine. The Propagandists seem to be enjoying a veritable "population explosion" these days. And their "aggrieved" protests are SO "Karl Rove" -- straight out of the RNC (or NSP) play book. :-(

great delivery, thorough, incisive, clear, and oozing integrity.

would that salon would publish THIS!!

i almost feel sorry for manjoo; not only will he have to deal with your quart of piss, the rest of us commenting unloaded a couple of gallons.

and given his past articles, he clearly does not have ANY of the details at his disposal with which to respond to any of your points. so he's painted himself into a corner, which was easy to see coming.

it will be interesting to await his response.

thanks again, joe; what a slamdunk! but then, he was never any real match for you, and he clearly could not see this coming.

saddest thing is, he may not even 'see' this coming, even when it flattens his face.

Well put. And even though I'm a regular reader, I don't mind that you keep making that point. Every time I read it, I'm reminded of the 2004 stench anew.

No one may ever address all of the problems in that election. Why would the press address the election, when they're still afraid to address the Downing Street Memo? They would never call our democracy into question, when they won't even call an abuse of power into question despite plain evidence from our closest ally.
completely off topic, but I think you should sell advertising on your blog. Quite a few folks are reading it. It's just i'd like to see you make a few bucks from your very good blog Joe. Help pay that skyhigh LA rent:) many lesser bloggers are making a few bucks, and kos is pulling in 20,000 a month. sheesh.

on topic. Not sure why manjoo is so quick to jump at any explanation besides fraud. These dueling statisticians are really pathetic. Statistics can be made to say anthing i guess, it's true. Bottom line for me, in ukraine exit polls off=fraud. In US=chatty dems. bullshit.
Manjoo did do some good reporting on evoting prior to the election. Sad to say he doesn't seem to have understood much of what he was reporting about. How could a journalist sound the warning about evoting, and then when a miraculous victory is achieved by the people who put those machines in play grasp at lame explanations is beyond me. People are so frightened somebody will call them a tinfoiler or something. pathetic.
Salon jumped the shark many moons ago. I've only checked in to read the war room blog in the last year. Not much else of interest to me there, and certainly not worth buying a subscription, IMO.
just for the record on the stats issue:

despite conventional, uh, 'wisdom,' statistics canNOT be made to say anything. the numbers are the numbers are the numbers. they represent what they represent.

however, that being said, all the action is in the 'be made to'. this is where human intent and motive come into play. if i have a list of statistical results, i'm ethically obligated to present these as they stand. however, some will pick and choose what they want to show. i've even seen scientists do this; oh, geez, that nagging little counterintuitive number over there, we'll just ignore it. typically these really are just bizarre outliers that no one can explain, so why try (though sometimes that's where all the interesting action lies). but in peer reviewed science, ignoring the big results, or misinterpreting them, virtually never passes muster.

i'm not a pollster, but from the pubs i've seen, they do less interpreting than just laying out the numbers. that's one reason all this incredible gymnastics to explain the discrepancyy away is so interesting; pretzel logic to show why the most reliable source of info does not jive with what in this equation should be considered the 'unknown' value. instead, as joe and others of us have pointed out, it's getting treated as the 'real' value, so some story has to be made up to explain away why the exit polls are in error. totally twisted. and again, as joe points out, these stories have no basis in reality, have not been empirically tested, and moreover are counterintuitive and fully illogical.

so the media ignore the issue because they assume that the public cannot understand the subtleties. and from manjoo's complete inability to do so, perhaps they are right.
Can we get this straight?

I provided a measuring stick. Mitofsky did the measuring.

I can offer an opinion as to what I think is the most likely explanation for the exit poll discrepancy, but without reanalysis, anything opinion I offer must remain "hazy" speculation. In my paper I called for reanalysis to be done, using my yardstick. Mitofsky presented one result of such an analysis at AAPOR, and used it to refute a specific fraud hypothesis advanced by USCV that "Bush strongholds have more vote count corruption". I believe some members of USCV still feel this is a viable inference. In my view, the test of this hypothesis was to ascertain whether there was a signficant linear correlation between Bush's share of the vote and my measure of bias. Mitofsky showed that while there was indeed such a correlation when the WPE was used as a measure of bias, this correlation vanished, as I predicted it might, when my index was used. In other words, the apparent correlation between the WPE and Bush's share of the vote was an artefact of the WPE.

This analysis does not in itself "prove" that response bias was the reason for the exit poll discrepancy. It simply, to my mind, refutes a particular fraud hypothesis. It also, to my mind, calls into question the likelihood that widespread, randomly distributed fraud took place on a scale to account for the entire "red shift". I now believe that fraud on this scale would tend to produce a real, significant correlation between my bias index and Bush's share of the vote - a correlation which is not present in the data.

Fraud may well still account for some of the "red shift"; however the lack of a significant correlation between bias and Bush's vote share, puts, in my opinion, fairly severe constraints on any hypothesised pattern of fraud that would be consistent with the data.
Anonymous write:

"despite conventional, uh, 'wisdom,' statistics canNOT be made to say anything. the numbers are the numbers are the numbers. they represent what they represent"

Hummmm, I might try that arrogant defense in traffic court or on a physics midterm....
dueling anons....

i highly recommend that you do just that, anon of the traffic court and physics test. appealing to the numbers on a physics test will no doubt bring you high marks, whereas attempting to reinterpret them might get you into serious trouble. same holds for traffic court, though the 'arrogant' comment would likely be used by the judge against you!

however, if you had read more carefully my caveat in the paragraph that followed that comment, you would understand that i am not being 'arrogant' about this point, but instead am aware that folks do in fact attempt to 'reinterpret' the numbers, making up all manner of stories about what the numbers themselves reference.

so if you go into traffic court and attempt to convince the judge that the numbers don't really reference your speed, but instead reference the temperature, you'll get into trouble. on the other hand, if you humbly admit that the numbers are the numbers, you might be granted some reprieve for your honesty and integrity. and if you try to make up fanciful stories about the numbers on your physics exam, god help you. the only way to pass a physics test is by admitting to the maxim: the numbers are the numbers are the numbers.

good luck with both!
Hey Blogger, I am conducting some research on various Web Investigator sites and ran across this page. Even though it wasn't "exactly" what I was searching for, I think the readers of "Manure from Manjoo: A response" still might be interested in reading our review of Web Investigator sites and other similar websites. As you know, the more information people have before spending their hard earned money, the better. Sorry for the self-promotion, just trying to get the word out while looking for more sites to review. On to the next -- Thanks!
As many links as you want!
Would The Thought Of Thousands Of Other Web Site's All Displaying Your Content,
Post a Comment

<< Home

This page is 

powered by Blogger. 

Isn't yours?

Image and video hosting by TinyPic

Image and video hosting by TinyPic

Image and video hosting by TinyPic