Our Man in Caracas

The Guardian continues with its stellar news reporting on Venezuela:

A survey for Datanalisis, a polling company, said 49% of likely voters would vote no and 39% would vote yes. A tracking poll by the opposition-linked Hinterlaces pollster predicted a technical tie.

Looks like Chávez is in trouble. But!:

It is difficult to make an electoral projections because the numbers of abstentions and undecided voters are high, the perfect ground for a surprise. The most probable is that there will be no surprise and Chavez will win 60 percent against 40 percent, but technically there is a possibility that it could happen, which is something that has not existed before.

That’s Luis Vicente Leon, pollster with local Venezuelan polling firm Datanalisis, yesterday.

To lend the Datanalisis poll additional credibility, Rory Carroll has compared it with the ‘opposition-linked’ poll Hinterlaces. As though Datanalisis were a more reliable source of statistical objectivity. Well, Datanalisis’s owner, Jose Antonio Gil Yepes is publicly anti-Chávez and has called for his assassination.

Carroll does include a caveat in his report:

Since his election in 1998, however, Chávez has never lost a vote and polls have a record of underestimating his support.

In the absence of any other information, you would think that the reason polls underestimated his support in previous elections is because, oh, I dunno, their sampling methodology just isn’t refined enough or maybe the country is too hot-bloodedly Latin to deliver meaningful statistics, rather than the real reason, which is that the polls are often big fat smelly fakes:

The international media has not always exercised due diligence in its reporting on polling data and elections in Venezuela,” said Weisbrot, who has authored papers on previous elections there.

“This opens up the possibility for the use of fake polling, as was done in the last (2004) referendum, to cast doubt on the results if the proposed constitutional reforms are approved,” he said.

In 2004, the influential U.S. polling firm Penn, Schoen, and Berland published fake exit polls on the day of the Presidential recall referendum, showing President Hugo Chávez losing by a 59-41 margin.(1) The actual results, which were certified by observer missions from the Organization of American States and the Atlanta-based Carter Center, showed the opposite, with Chávez winning by a margin of 58 to 41 percent.(2)

The fake exit polls were not the only dubious polls that plagued the last referendum. Most of the pre-election polls in 2004 showed the race “too close to call.” Although these were conducted by opposition pollsters, most of the international media accepted them in their reporting. As CEPR demonstrated at the time, it is extremely unlikely that a properly conducted poll could have shown a result that was “too close to call.”

I can certainly understand why Rory Carroll might report in this way. What I don’t understand is why he needs to be in Caracas to do it. You could sit in a cublicle in Swindon and write this stuff.


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

I on Twitter

November 2007
« Oct   Dec »

%d bloggers like this: