ON June 4, 2010, the Stabroek News (SN) published a March 2010 poll conducted by Barbados-based Caribbean Development Research Services Inc (CADRES) which placed the governing PPP/C as the party commanding the largest single bloc of voters in Guyana.
The SN article went on to quote a CADRES press release which stated, “In the final analysis, however, the PPP/Civic is still leading, the PNCR is still the second most popular party; however the AFC would have been in a position to hold the balance of power if an election were called in March this year.”
According to CADRES, approximately 1,000 Guyanese voters were randomly selected and interviewed. In each instance the sample was proportionate to the percentage of the population that actually lives in the respective administrative region. The margin of error associated with such a survey is +/- 5% and as a result, the results could be said to be a “reasonable indicator” of public opinion in Guyana at the time it was conducted.
In defining their methodology for conducting polls elsewhere, CADRES state that they try to obtain samples of at least 1000 participants and interviews are done “face to face”. It is strange that the 1000 participants, the team of interviewers, the University and/or other NGOs, that must have assisted, carried out this mammoth exercise in Guyana without even “them boys” noticing and/or saying anything about it.
Guyana is not a country where opinion polls are conducted and published with any degree of regularity. Guyanese are apt to declare support for political leaders who present themselves in their domain as in the case of CN Sharma who was more than once polled as the most popular political leader in Guyana. Any poll that says otherwise must be discredited as the Guyanese population gets its regular dose of “Sharmanese” on CNS Channel 6 thus ensuring sustained popularity. Such declarations of popularity, however, have never translated into votes on elections day.
It is not usual for a political poll to be made public so far outside of the ‘political season’ and it is glaringly obvious that the results of this poll were not published by any other media house in Guyana despite the claim that the information was obtained via a CADRES press release.  What is remarkable is that no other media house in Guyana appears to have received the press release. This led to the obvious conclusion that there was some kind of collusion going on with the persons who commissioned the poll and the Stabroek News.
Dr. Randy Persaud, in a series of letters to the SN, sought to query the circumstances surrounding the publication of this poll but was met with a series of evasive answers by the newspaper until June 08, 2010 when the SN published an article claiming that Floyd Haynes of the Washington-based Newton Group had commissioned the recent CADRES poll and that Haynes is married to former AFC MP Chantelle Smith who, incidentally, was also a former employee of the Stabroek News. This new revelation brings into question the CADRES’ Director of Research Peter Wickham’s claim that none of the political parties or the political party leaders had anything to do with the poll.
CADRES, under the direction of political scientist Peter Wickham, conducts social and political surveys and related research in the Caribbean. It also offers Political Campaign Management Services. Among its major projects/consultancies are public opinion polls in Antigua, Barbados, Bermuda, British Virgin Islands (Tortola), St. Kitts, St. Vincent, St. Lucia, Trinidad and Tobago, Turks and Caicos Islands and Guyana, since 1993. Its polls have been commissioned by newspapers, political parties and governments.
No mention is made of Grenada where CADRES got its poll wrong. It predicted a victory for Dr. Mitchell’s party which was defeated 11-4 by the main opposition, National Democratic Congress (NDC) in general elections.
In Dominica, the DLP issued a Press Release in December 2009 about a recent CADRES poll which states “We are of the view that this is not scientific and is an outrageous attempt to influence the thinking and choice of Dominica voters and should be rejected out of hand. This constitutes another blatant effort to manipulate the electorate and to tarnish our democracy and the process of free and fair elections. We wish to advise our people to exercise extreme vigilance and not fall for another devise that is now added unto the multiple corruptions, chartering of flights and the Prime Minister’s thinking that he is above the law and beyond any semblance of decency.”
In Barbados, Wickham’s articles were banned by the Barbados Nation. Former Barbados Prime Minister, Owen Arthur, describes Peter Wickam’s poll in Barbados as part of a deliberate attempt to manipulate public opinion. According to bajan.wordpress.com CADRES polls in Barbados were biased as “Peter Wickham was/is a member of the DLP, is on very friendly terms with the PM and is known to be a strong supporter of the government. So please could we not put aside the bogus outcomes of any CADRES poll right now. Peter Wickham is just what he is, part of the Hartley Henry gang, talking up the DLP in the face of a mountain of domestic problems.”
Finally, although CADRES usually makes claims about the accuracy of their polls, polls in Antigua, Grenada and St Lucia predicted victories for the ruling parties. In these elections the ruling party lost at the polls contrary to the predictions of victory made by CADRES.
Now, the significance of these alliances becomes apparent and Dr. Randy Persaud’s point of bias and unprofessional conduct becomes an issue. Examples such as these prompted media monitors to establish a new category of concern called Media Effects.
Media Effects is a special category used for those occasions which are not possible to monitor neither quantitatively or qualitatively, as they simply do not fall into any of the established categories.
In other words, these are all stories or items presented with a specific intent to manipulate or affect the public. The reason for creating this category was the obvious absence of basic journalistic principles and standards. Each story which does not meet the basic requirements for a non-biased story falls into the category of Media Effects.
The main criteria for such an evaluation are set as those stories missing: relevance, exactness, transparency, matter-of-fact, balance, variety, timeliness and clarity.
No one can doubt that scientific polls provide the best source of information about public opinion. They are valuable tools for journalists and can serve as the basis for accurate, informative news stories. However, the journalist looking at a set of poll numbers is obligated to ask some questions of the pollster before reporting any results.
One of the most important things to know is who paid for the survey because that tells you and your readers who thought these topics are important enough to spend money finding out what people think. In the case of CADRES, the fees released for a previous poll was more than US$20,000 and US$8,000 was paid in advance. The important issue for a journalist is whether the motive for doing the poll creates such serious doubts about the validity of the results that the numbers should not be publicized. Private polls conducted for a political campaign are often unsuited for publication. These polls are conducted solely to help the candidate win – and for no other reason.
In a press statement, the PPP said that it wished to draw attention to the resemblance between this poll, and what it called the “fraudulent” one made public by the AFC in 2006 which was conducted by Dick Morris. The statement went on to say that at a press conference, Morris had claimed that the AFC had overtaken the PNCR-1G in terms of voter support, although he and th
e AFC were ‘dead wrong’ since the PPP won 54.5% of the votes.
The PPP said the Morris poll had also predicted that the AFC would hold the balance of power in Parliament, a prediction which was wrong.
A check of its website reveals that CADRES provides Political Management services and that “Political campaign management services were conducted in Guyana in 2006 and included such services as strategy formulation; constituency reports; political advertisements and materials.”
It is known in the industry that some political campaigns and special-interest groups have used a technique called “push polls” to spread rumours and even outright lies about opponents. These efforts are not polls, but political manipulation trying to hide behind the smokescreen of a public opinion survey. “Push polls” are unethical with the only objective of disseminating false or misleading information. Such polls have been condemned by professional polling organizations.
In the face of all the evidence at our disposal, it is natural to conclude that there was some amount of bias in the conduct of this poll, especially since CADRES was involved in Guyana as a Campaign Manager in 2006.
From the results of the polls we can draw our own conclusions as to the party which benefitted from the political Campaign Management services offered by CADRES and wonder whether they are still on retainer.
We must commend those responsible publishers who received this information and refused to publish same and by the same token condemn those irresponsible publishers who choose not to make the elementary checks required of their position in service to the nation.
In other countries, privately owned national newspapers endorse political parties before a general election. If the Stabroek News wishes to endorse a political party it should follow the Mirror and have the courage to come out and say so in no uncertain terms rather than rely on subterfuge and collusion in the publication of push polls.
Any newspaper that wishes to endorse a political party should do so as soon as possible so that readers are provided with a context with which to analyze what is being published. In the final analysis, true objectivity is only possible when bias is declared. The experienced and more level headed staffers at Stabroek News must understand that it is not ethical to manipulate and deceive under the cloak of declared objectivity.
(Sources: bajan.wordpress.com, cdsonline, globeinter.org and wikipedia)