Thad blogs favorably on recent pieces by Jeff Greenfield and Chris Hitchens (see below).

There are many kinds of “democratic” processes out there, and it isn’t correct to presume that a secret ballot cast in private is the only set of voting institutions that merit the label “democratic.” For example, to advocates of “deliberative” democracy, no process which relies exclusively on an individualistic, atomized decision making procedure can be called “democratic.” Democracy requires some level of consensus in the community, and, to these theorists, we build consensus via conversation and exchange of ideas, not through one-way political advertisements.

And to the defender of Iowa, the caucus process displays these features in spades. Citizens publicly defend their preferences. They publicly try to convince their friends and neighbors to change their minds. And they “cast” their ballots by grouping themselves with others.

My main concern with Iowa is not the caucus process per se, but the unrepresentativeness of the population and the low level of participation. But precisely the features that Hithchens and Greenfield hate–the public nature of the discussion and vote–are features that I find rather appealing.

Andrew Gelman’s blog is a nice place to keep up to date on good ways to present and discuss statistical data. Some of the discussion can be complex, but Gelman has made it one of his missions to publicize simple and clear presentations of statistical material (and, watch out academics!, criticize bad presentations).

A nice example of presenting voting results via maps is contained here.

It’s the end of the holidays and the beginning of the primary season. And this is the first, but sadly not the last, posting I’ll make about examples of innumeracy among journalists.

In yesterday’s coverage, Judy was asked by Rey Suarez about the most recent Des Moines Register “Iowa Poll”, which showed Barack Obama 7 points ahead of Hillary Clinton. Since five other polls showed different results (the race too close to call), Suarez asked Woodruff: “Is this poll an outlier?”

Woodruff went on to restate how “important” this single poll result was, and peddled some stock quotes from Mark Penn and Harrison Hickman. She never showed any sense that she understood what an “outlier” was, or how this piece of information may have helped the viewers understand the status of the Iowa contest.

Tonight, in responding to Woodruff’s breathless commentary about the number of advertisements in Iowa, Gwen Ifill asked “This must be costing the candidate’s a lot of money.”

“Yes,” Woodruff replied,” but most if it is Democratic. The Democrats have spent nearly 80 million dollars in Iowa, compared to 18 million in 2004. So they are spending magnitudes and magnitudes more.”

Should we just chalk this up to hyperbole? Is it just picky for me to point out that a “magnitude” means something is ten times larger, and “magnitudes” means 100 times, 1000 times, or larger? (They are spending four times more. But I guess that is a lot less impressive that “magnitudes” more.”)

This is fodder for bar conversations among statistically inclined types, but there is an important undercurrent: to many educators, including me, “numeracy” is an important part of the educational curriculum and is too often ignored, particularly in college curricula. Numeracy (“mastery of the basic symbols and operations of arithmetic”) is a purposeful analogue to “literacy,” and, many argue, is just as critical in modern life as are reading and writing. And certainly, any political journalist who is going to make use of polling data must understand basic statistics.

Jeffrey Greenfield also has a piece in Slate and he notes that it is the Democrats who have the democratic election problem in Iowa, not the Republicans.

The original purpose of the [Iowa] caucuses—to conduct party business and to talk over local concerns—became completely overwhelmed by the presidential frenzy for which they’re so ill-suited. As Drake University professor Dennis Goldford notes, “The presidential preference just began as something piggybacking on an ordinary set of party functions, and it’s been blown way out of proportion.”

Beyond that, the Democratic Party’s caucus method requires not 10 to 15 minutes at a polling place, but two or three hours in a school lunchroom or library. This is why turnout—measured by eligible voters—ran under 6 percent in 2000 (the last time both parties held contests).

The Republican Party, by contrast, has recognized that the change in function, from local party business to presidential contest, requires a change in form. The GOP caucus process is straightforward and simple: You show up, perhaps listen to appeals from candidate’s supporters, and then write the name of your choice on a blank piece of paper and drop it into a box. The results are phoned into headquarters and tabulated. That’s it—one person, one vote; the candidate with the most votes wins.

But the Democrats have a totally different thing going on; one that discards at least two key elements of an open, fair system: the secret ballot and the one-person-one-vote principle. When you show up at a Democratic caucus, you and your fellow participants divide up into different corners of a room, based on who you are for. You don’t submit a secret ballot; you stand up to be publicly counted. What if you’re in a union and want to pick someone your union hasn’t endorsed, and your shop steward is there, watching you from across the room? Or the person who holds your mortgage? Or your spouse? Tough.

The article explains even more silliness in the Democratic Party rules. It is fun, though kinda sad, reading.

The fun of reading Christopher Hitchens is that he is so mild, so demur, so lacking in scathing prose (just as I sometimes lack sarcasm). Here is what he wrote today in Slate.com about the Iowa Caucuses (the whole story is linked above).

It is quite astonishing to see with what deadpan and neutral a tone our press and television report the open corruption—and the flagrantly anti-democratic character—of the Iowa caucuses. It’s not enough that we have to read of inducements openly offered to potential supporters—I almost said “voters”—even if these mini-bribes only take the form of “platters of sandwiches” and “novelty items” (I am quoting from Sunday’s New York Times). It’s also that campaign aides are showing up at Iowan homes “with DVD’s that [explain] how the caucuses work.” Nobody needs a DVD to understand one-person-one-vote, a level playing field, and a secret ballot. The DVD and the other gifts and goodies (Sen. Barack Obama is promising free baby-sitting on Thursday) are required precisely because none of those conditions applies in Iowa. In a genuine democratic process, these Tammany tactics would long ago have been declared illegal. But this is not a democratic process, and besides, as my old friend Michael Kinsley used to say about Washington, the scandal is never about what’s illegal. It’s about what’s legal.

He then notes that:

It’s only when you read an honest reporter like Dan Balz that you appreciate the depth and extent of the fraud that is being practiced on us all. “In a primary,” as he put it, “voters quietly fill out their ballots and leave. In the caucuses, they are required to come and stay for several hours, and there are no secret ballots. In the presence of friends, neighbors and occasionally strangers, Iowa Democrats vote with their feet, by raising their hands and moving to different parts of the room to signify their support for one candidate or another. … [F]or Democrats, it is not a one-person, one-vote system. … Inducements are allowed; bribes are not.” One has to love that last sentence.

The winner of the caucus will be the person who gets a tiny sliver of a tiny sliver of participants. And alas, most people reading this will hang on the media’s coverage of the caucus anyway.

The Kenyan elections have turned out to be quite a fiasco and a tragic one at that. The problems started when it looked as though the opposition leader would win, then that win dissipated.

As the Telegraph initially reported: At noon yesterday, the electoral commission gave the challenger, Raila Odinga, a four per cent lead over the incumbent, Mwai Kibaki, after results from three-quarters of the 210 constituencies. But almost 48 hours after polls had closed, hold-ups to the remaining 51 counts drew sharp criticism. Reuters noted that: Kenyan election officials called a halt to vote tallies on Saturday, leaving the result of a cliff-hanger presidential vote in the balance. Electoral Commission of Kenya (ECK) officials told journalists to come back on Sunday for further tallies. With 86 percent of constituency results in, President Mwai Kibaki and opposition leader Raila Odinga were neck-and-neck.

Such delays of often viewed as problematic because they give time for shenanigans to occur. And election observers suggest that vote counting shenanigans likely did occur. As the New York Times reported:

It had been predicted that this election would be close, and the final results had Mr. Kibaki winning by a sliver, 46 to 44 percent. But that gap may have included thousands of invalid votes. The European Union said its observers in one constituency last week witnessed election officials announce that President Kibaki had won 50,145 votes, but on Sunday the election commission increased those same results to 75,261 votes.

“The election commission has not succeeded in establishing the credibility of the tallying process,” said Alexander Graf Lambsdorff, the chief European observer. One Western ambassador said that Western diplomats tried for hours on Sunday to persuade the election commission to do a recount of the vote figures using original results but that the commission refused. “This was rigged,” the ambassador said.

The election commission acknowledged that there were irregularities but said that it was not their job to address them. The opposition, said the chairman, Samuel Kivuitu, “can goNew York Times to the courts.” The opposition has not indicated whether it would contest the results in Kenya’s courts, which are notoriously slow and corrupt.

As the New York Times, NPR, and others reported today, the violence in Kenya between supporters and opponents of the president have become fierce. Numerous people died today in clashes, including 300 people in a church.

USA Today has a big story today on voter rolls.

Five years after passage of a federal law to create electronic registration databases to deter voter fraud, the new technology is posing hurdles that could disenfranchise thousands of legal voters, a USA TODAY examination finds.

From Florida to Washington, voters have been challenged because names or numbers on their registration forms did not exactly match other government databases, such as Social Security and motor vehicle agencies. “We know that eligible people have been thrown off the rolls,” says Justin Levitt, a lawyer with the Brennan Center for Justice at New York University School of Law.

The databases are only as good as the information fed into them by applicants and election officials. That can lead to human errors as well as variations from state to state. Colorado, for instance, knocked nearly 20% of its voters off the rolls between the 2004 and 2006 elections. Arkansas purged 3%, according to Election Assistance Commission data.

They also have a related story that you can link to here.

In our forthcoming book on election fraud, we have a chapter we co-wrote with Rod Kiewiet on “low-tech” election fraud, such as creating disturbances at polling places in areas of high concentration of “bad” voters in order to disrupt or halt voting (thus lowering turnout in one party’s stronghold area). It seems as though the Pakistani intelligence services may have hacked into Rod’s computer and read it.

“Where an opposing candidate is strong in an area, they [supporters of President Pervez Musharraf ] have planned to create a conflict at the polling station, even killing people if necessary, to stop polls at least three to four hours,” [according to a report written by supporters of Bhutto].

The report also accused the government of planning to tamper with ballots and voter lists, intimidate opposition candidates and misuse U.S.-made equipment to monitor communications of opponents.

We never considered the idea of killing people, but low tech election fraud is clearly something that people think is a worthwhile activity.

This is a fascinating article on electronic voting, ballot layout, and voter confusion.

The Associated Press has a story (the link above is to the version from the International Herald Tribune) where

South Carolina Republicans got a bogus holiday greeting card this week, purported to be from White House hopeful Mitt Romney, that cites some controversial passages of the Book of Mormon. “We wish you and your family a happy holiday season and a joyful New Year. The Romney family,” the card says.

One question that the article raises but does not answer is whether this constitutes election fraud (as opposed to some other sort of fraud). Note that the quoted sections do not engage in intimidation (e.g., vote for Romney and you will be fired from your job or harmed) nor does the mailing steal votes or offend voter registration issues.

I bring this up not to minimize the tactic, which is clearly wrong and, as one consultant points out, probably a tactical mistake, but to note that election fraud can be a slippery issue that is very hard to pin down at points. It also raises a question of how we delineate between campaign “frauds”, such as this dirty trick, and what constitutes a fraud on an actual election.