It is difficult not to get caught up in the politics as we enter these last few days before August 10th if you have even the slightest interest in politics. Certainly, you can't open a newspaper without it hitting you squarely in the face. And here (again) is my central complaint.
Look, I don't know how many times I have to say it: I genuinely don't care what your political, religious or other preferences are. As far as I am concerned, while I reserve my right to disagree with you on anything, I absolutely agree that you have an equal right to disagree with me and we can meet on the battlefield of ideas.
But I do insist that you should not be afraid to declare your bias, especially in circumstances where you are a newspaper and there are readers who are depending on you to report the news accurately. So, when I read this morning's Guardian and saw in big headlines on page 6 "DEAD HEAT IN BARATARIA/SAN JUAN" and again on page 7 "PNM SEEN AS PROBLEM SOLVER" I decided to read the article which was purporting to report on a poll by Louis Bertrand of Hill and Associates. The article together with accompanying tables was most instructive.
First of all, the poll was conducted with only 200 respondents in that constituency which has more than 24,000 registered voters. Okay, you might say, but pollsters have their own methods of extrapolating coming events from relatively small samples. Okay. But everyone knows that this election is going to be heavily charged with racial tensions and that generally speaking if you are African you are voting PNM and if you are Indian you are voting UNC. So in order to be able to assess the polls findings, shouldn't we know exactly how many of the 200 respondents were Indian/African? Because, for example, if 150 of the respondents were Black and only 50 were Indian then I would say that the poll really doesn't show a dead heat but in fact it predicts a UNC victory. Put another way, the report in the Grand Old Lady of St. Vincent Street is missing some key information.
Again, while I am no pollster the published results in the newspaper don't seem to line up with the information in the tables. For example, 29 percent of the interviewees are reported as saying that they think that Saddam Hosein (the UNC candidate) was doing a good job versus only 12 percent thinking the same thing of Jason Williams (the PNM candidate). But 14 percent had an unfavourable opinion of Mr. Hosein versus only 9 percent for Mr. Williams. To which I can only say what?! You do see the obvious dichotomy there, don't you? Then there is a table (table 61) which reports that 40 percent of the respondents think that UNC will do a better job of all the problems that face the constituency than 33 percent who think that PNM will do a better job.
Then buried in the article is the quote "Race, as a factor in voting intentions, continues to be important in the constituency." What did I just say a little earlier? And then the article goes on to say that 59 percent of Indo-Trinis intend to vote for the UNC versus 9 percent for the PNM. But 66 percent of Afro-Trinis intend to vote for the PNM versus only 6 percent for the UNC. But all available evidence suggests that PNM supporters are generally more dissatisfied with their party than are the UNC supporters. Certainly, you hear more complaints about the PNM coming from their supporters than you hear from UNC supporters about their party. So? Is what the poll is saying really true?
You realize, of course, that either there is something wrong with the editor who allowed this report onto the newspaper's pages or there is something wrong with the pollster. Because the published information is simply too full of contradictions to come to the conclusions that there is a dead heat in the constituency or that PNM is seen as a problem solver. No. It just isn't making sense.
But, the headlines make a lot of sense if you realize that most people only read the headlines and you think that the Guardian is trying to prop up the PNM vote.
No comments:
Post a Comment