Thursday, January 10, 2008

Crocodile tears vs. compromised code


The topics of Hillary Clinton's "upset" victory in New Hampshire, and the polls which predicted it wrong, were the focus of much discussion on the 24 hour news cycle blabfest talks shows. From my perspective, the arguments fell into three categories, in which specious reasoning resonated.

1. The polls, right or wrong. The press, and the punditry in particular, jumped all over the "inaccurate polls." If we assume that the polls simply got it wrong, then we have to acknowledge that all the polling companies got it wrong at the same time, and in the same way. Each of the companies predicted correct outcomes for candidates, with the exception of the number of votes Hillary Clinton would receive. If there was aberrant, unpredictable behavior in New Hampshire voting, then none of the polling firms recognized it. In some ways, a true aberration would likely escape the attention of all polling firms in the same way. That has sent the journalists scurrying for explanations as to what the aberrant voting behavior was, who those voters were, and what motivated them. Questions likely never to be answered. However, very few journalists have even raised the issue that maybe the pre-election polls, and notoriously accurate exit polls, may have been correct. Even after voting malfeasance has been demonstrated in the last Presidential election in Ohio, where the last occurence of "incorrect exit polls" became news, even after a recent major article in the New York Times Magazine cast doubt on the accuracy of electronic voting machines, even after it's been proven that Diebold machines (the machines used in New Hampshire) are easily hacked, even as new questions surface about LHS Associates, the company which programs New Hampshire (and all Connecticut) electronic voting machines, even after Republican election workers in New Hampshire have been indicted for phone jamming, journalists are loathe to consider even the possibility of voting machine error or fraud.

2. The mysterious reasons for Hillary Clinton's last-minute surge. Everyone including Hillary was surprised at her victory in New Hampshire. Why? Because polls predicted she would run a distant second. So the punditry has been tied in knots trying to figure out what happened. Here are the theories I've heard.

- Clinton cried on camera. Assumption: voters finally discovered Clinton's "human side."
- Clinton talked for two days about the economy. Assumption: it's the economy stupid.
- Bill Clinton scolded Barack Obama. Assumption: Voters realized Obama is not such a nice fellow.
- New Hampshire voters decided to prove the press wrong. Assumption: People from the land of "live free or die" were not about to allow the press to anoint Obama as the presumed candidate.

All of these explanations can be lumped into the "grasping at straws" category from a press drowning in its own humiliation at getting it wrong (Karl Rove thinks they're right). Still, there's one more unlikely explanation, as implausible as these, but actually better able to be proven than any of the above: someone messed with the voting machines.

Yet this explanation - voter machine fraud - was either not even considered, or scoffed at by pundits, including our hero Keith Olbermann (his shrugging dismissal of this alternative explanation is at the end of this segment):



3. Horoscope or journalism. There's a longstanding question as to whether polls should be treated as "news." They certainly aren't fact, but projections, and even the polling firms will tell you that. Newspapers, new shows, and particularly pundits treat polls as if they are gospel truth. In doing so, some think that news reporters actually shape the news by affecting the outcome of elections. Newspapers might as well print, as fact, the horoscope on the front page - or the weather predictions for a month in the future - as they do campaign polls.

These same news outlets have been quick to jump on the polls as being "wrong," but I haven't found a headline yet that reads: We Were Wrong. Much easier to place blame on someone else.

It's what The Hartford Courant did this morning in an editorial scolding the pundits and pollsters. This from a newspaper who as late as Tuesday morning January 8, featured a front page headline that read: Clinton on Shaky Ground As Obama Surges Ahead, as if it were fact, and not a reflection of polling (a projection, a prediction, a guess).

The press have been wrong so often in the last eight years that they should be ashamed to point a finger elsewhere. As Tom Brokaw suggested to Chris Matthews, journalism needs to ignore polls and concentrate on other campaign matters, like a candidate's record, campaign tactics, claims and counterclaims. Seems like there's a better role for journalists to play than to ride on the campaign bus, listening to the same stump speech six or eight times a day, and getting chummy with a candidate (except of course in the case of Hillary Clinton).

No comments: