Monday, December 17, 2007

Iowa Caucus Troubles

Oh, boy, what a mess. I've heard plenty of whining from all quarters about how bad the Iowa Caucuses are for democracy and whatnot: we're an overwhelmingly (and hence not nationally representative) white, the caucus itself isn't really a democratic process, Iowans are idiots. I have always been, to varying degrees, sympathetic to all of these arguments. (Well, not terribly sympathetic to the idiot charge: I am biased.)

And, then, via Mickey Kaus on Slate, comes trouble. First of all, Kaus says, "The TV networks are screwing around with the already-absurd Iowa caucuses again, using an 'entrance poll' of only 40 precincts (out of more than 3,500) that threatens to manufacture a misleading result. Ah, but it's all justified because of the valuable information the network poll will gather!" (Emphasis in original). He continues, "The networks wouldn't have to resort to a questionable 'entrance' poll if Iowans voted at normal hours using, say, easily-countable ballots. But that's not the Iowa way." Read on for the whole backstory, or skip down to my interpretation.

Kaus is generally one of the Slate features I generally avoid -- I'm often sympathetic of some of the highly cynical points he makes, but I tend to choke on his vitriol. But his arguments, if true, were distressing enough that I read with interest. Most damningly, however, he provided a link to a piece by Slate correspondent William Saletan, who writes the "Human Nature" feature on that site. Describing the article as "seminal," I followed the link to Saletan's "article on the epic 1988 caucus debacle." The title from the Saletan piece? "The Phantom Poll Booth: Who won the 1988 Iowa Democratic caucuses? We'll never know." (I was only 5 during the debacle -- I didn't even know that there had been such a mess.)

The 1988 piece is presented as a sidebar, reproduced from the June 1988 American Politics, to a 2004 post, "The Vanishing: If you liked the Florida recount, you'll love the Iowa caucuses," written by Saletan and Slate intern Matt Schiller. Things aren't getting any better, apparently:

To create a dangerously high risk that the winner of the delegate count isn't the winner of the raw vote, you need two things: a big field, so that there will be plenty of nonviable groups to redistribute at the precinct caucuses, and a close race.

...

This year, the two risk factors have returned with a vengeance. The field is bigger than in 1988, and the race is closer. The latest Iowa polls have the top four candidates—Dean, Gephardt, Kerry, and Edwards—within the margin of error. [A media consortium called the News Election Service], which tried to count the raw vote in 1988, is gone. Here's the system the media have created to replace the NES: Nothing. Plenty of reporters will attend caucuses, but nobody is systematically reporting the raw vote, or even the realigned vote. Some folks at the TV networks seem to think the Associated Press is reporting the raw vote. That's news to the AP.

For better or worse, all of this presupposes you have something of an understanding of how the caucuses work. Don't worry if you're not an expert: it doesn't appear that anyone is. (Or perhaps that's a good reason to soil your trousers. On reflection, I pick the former.)

Interpretation
For those of you who don't have half an hour to dispose upon your democracy (shame!), or trust me enough (horror!) to be interested in my read on the situation, well, here goes. First, a quick primer on how the caucuses work, with reporting measures thrown in to taste. This is hardly an authoritarian or complete description; I'm just touching on the points that are implicated in Kaus and Saletan's writing. And note, this only applies to the Dem's. The GOP has their own form of a caucus, which is strikingly like a primary.

Alright. January 3, 7 P.M. Some Concerned Citizens (the eligibility requirements are an entire 'nother post) gather at their designated caucus spot. At a few of the zillion local precincts, the press will take entrance polls of caucus goers' first choices -- before the caucus begins. Right. The precinct captain tells people to go stand in the corner of their respective candidate: Hillary folks in one corner, Obama supporters over there, Edwards' over here, Bidens' in the fourth corner, Dodds' here between Hillarys' and Edwards', ad nauseam. It is at this point, to my understanding, that the former NES (now defunct, see above), was instructed to take vote counts. Candidates groups not meeting viability are told to realign. Viability is figured as 15% of precinct attendance. During realignment, previously nonviable candidate support groups can achieve viability by bringing on supporters from other nonviable candidates, or by sniping from viable candidates. I saw this happen in 2004: when Dean and Clark all failed to meet viability, Dean supporters sat out for the most part, and Gephardt's group offered a few supporters to Clark, ensuring that Dean wouldn't get a delegate.

Confused yet? Gephardt's group was able to offered two or three supporters for a Clark delegate because of the way that rounding works. Imagine division with a remainder, if you would. In my particular precinct, for example, we were allotted six delegates in 2004. Given to the number of people who showed up -- 80 at my Lone Tree, Iowa, precinct -- and the 15% rule, 12 persons were required for viability. However, every 14 persons in a particular group (80 attendees / 6 delegates = 13 1/3) provided a delegate. In theory, therefore, it would be possible to be viable but not earn a delegate. My head is starting to hurt, too.

Dean's group, if I remember, was one supporter short of viability -- 11. Another supporter would have allowed them viability, and the chance to snipe supporters from other groups. If you remember, Iowa was a fight between Dean and Gephardt -- so Gephardt would want anyone else to get a delegate before Dean. Therefore, they offered a few strategic bodies to Clark, assuring him viability and a delegate. Without those bodies, Clark supporters might have gone over to Dean's camp, giving him a delegate.

What's all of this have to do with division and remainders, you ask dear reader? Gephardt had sixteen supporters. Let's practice division: 16/14= 1 remainder 2. Those two extra bodies in the Gephardt camp were pretty much worthless: they weren't going to give Dick an extra 1/7 delegate. But they could use those two bodies to shore up Clark -- who turned out to be exactly the also-ran Dick's crew calculated them to be -- and ensure that a measly three bodies didn't deflect from Clark to Dean. (As an aside, Clark earned exactly three statewide delegates in 2004, in Iowa. The Clark county delegate from Lone Tree in 2004 was one of three in Johnson County (same county as Iowa City) and one of 15 statewide. Confused, still?)

Now, this isn't even the problem! This is serious enough to require both boldface and italics.

As you can begin to imagine, the whole caucus process takes a long bloody time, the antithesis of political news coverage. Rewind to the beginning of the caucus -- the entrance poll. My best read of Kaus and Saletan is that the entrance poll numbers are what will be reported on the news outlets -- entrance poll numbers that tally caucus goers' first choices, and poll numbers taken from only 40 precincts (of a total 3,562 Democratic and Republican precincts).

Implications? Well, it would seem like no one need caucus unless they happen to be at one of the 40 precincts where there is entrance polling taking place. And, then, who do declare your support for? Imagine a caucus-goer who supports Sen. Dodd, but who figures Dodd can't win. Not knowing that her vote really counts incredibly disproportionately highly, this caucus-goer might declare her support for her second choice (and top-tier, per MSM) candidate, Sen. Edwards. Arg!!! Not only is her vote subverted (well, not really a vote), but well, everyone who caucuses elsewhere need not apply anyway. From Saletan:
In the last days before the caucuses, buoyed by the Des Moines Register's en­dorsement, [Paul] Simon aides echoed the Bab­bitt campaign's mocking line on Gary Hart: "Let the media decide." On Febru­ary 8, that's what happened.
Looks like that's set to happen, again. Please -- someone tell me that I'm wrong about this. I'm beginning to make a Kauslike use of bold and italics. Save me!

[Edit 071217@2109: Readability edit.]
[Edit 071217@2146: Readability edit again.]

1 comment:

Lacey said...

Wow, I found all of this extremely interesting, and of course had to read through a few of the articles you mentioned. I had no idea about all this, but I'm actually going to a meeting tomorrow night about the caucuses so hopefully I'll be even more informed on procedures.

As to commenting on this information, though, I have a few thoughts.

1) I find it hard to believe that someone is going to attend a caucus and get counted in the beginning, and then change their vote based on the media numbers within a half-hour. After all, the counting from the media (it sounds like) starts at 6:30pm, and the doors close for caucus-goers to be a counted number at 7pm, so the idea that someone would use their Blackberry cell phone (as was mentioned in one of the articles) to see the press's numbers and change their vote, strikes me as hard to believe. I would imagine the people attending the caucus would be too caught-up in the moment/speeches/persuasion statements to be trying to catch what the press's numbers are reporting.

2) Iowa's stance of not having a primary versus a caucus makes sense to me. Considering Iowa wants to keep any press coverage they can get due to Iowa not receiving a whole lot of attention, especially nationally, I think the fight for keeping the caucus the way it is is understandable, but perhaps not ethical due to inaccuracies.

3) Another thought: Are we to assume that the ones attending the caucus are the most knowledgeable on the candidates, and if so, is debating/persuading/rearranging necessarily a bad thing? Can one assume that the speeches going on during the caucus are educating people and therefore leading them to a better informed decision on who would make the best president? Could caucusing be enabling someone to change their mind to a candidate more aligned with their core values as a voter?

Sounds like to me the best solution might be for the media (who understandably does not want to wait for the results by having to listen to people bicker) should perhaps just ask each caucus place to call in when they're all done with their numbers at the end of the caucus. However, I'm sure this would be against the media's desire for high ratings/getting out the news ASAP, whether accurate or not? I'm not sure if this is the case, but that's what I'm gathering from this.