december 01, 2007

Varning för manipulationer


Igår avslutatade Chávez kampanjen i Caracas

Läget är nervöst inför söndagens folkomröstning. Jag tror att Chávez och Ja-sidan borde vinna med minst 55% av rösterna. Opinionsmätningarna har gett spridda resultat och är oftast gjorda av partiska intressen. Avgörande blir hur de olika sidorna kan mobilisera på valdagen.

Mycket talar för att oppositionen komer avfärda valresultatet och iscensätta en plan för att försätta Venezuela i kaos. Att Venezuelas har kanske världens säkraste valsystem spelar ingen roll. Måndagen kommer antagligen att bli våldsam. Frågan är hur våldsam. Ingenting kan uteslutas. En stor varning måste utfärdas för hur massmedia kommer att beskriva händelseförloppet.

Se också detta:

WASHINGTON, D.C. - In an interview made available Wednesday, November 28, the head of a widely-cited polling firm contradicted his firm's own findings that had suggested a majority of likely voters would vote against proposed constitutional changes favored by Venezuelan President Hugo Chávez.

In an interview with Reuters posted online yesterday, Luis Vicente Leon, a pollster for Datanalisis stated, "The most probable [projection] is that there will be no surprise and Chavez will win 60 percent against 40 percent."

Datanalisis' prior survey has been the most widely cited poll in the international media in the run-up to Sunday's referendum. The poll, conducted on behalf of private businesses, and reported in the media on November 24, reported that "about 49 percent of likely voters oppose the reforms while 39 percent favor."[1]

Vicente Leon's statements directly contradict the results of the poll. If the 60/40 projection is correct, CEPR calculated that the odds of obtaining the prior poll result - i.e. 49 percent of respondents saying no -- would be one in two billion trillion.

"As a statistical matter, Mr. Vicente Leon's remarks can be considered an admission that his previous poll was not valid," said CEPR Co-Director Mark Weisbrot.

CEPR had warned just yesterday that dubious or fake polls might be used to influence public perceptions during the election.

1 kommentarer:

Anonym sa...

hvorfor ikke:)