Beware of opinion polls

Many surveys and much of market research that is conducted appears to have hidden agenda and distinct purpose in mind.

I have always been rather suspicious of opinion polls. The business of fashioning truly effective survey questions is not easy. Sampling errors is a built-in and unavoidable feature of all popular polls. Many of the surveys and much of the market research that is conducted appears to have a hidden agenda and distinct purpose in mind, such as the endorsement of a particular electoral candidate or the promotion of a certain product at the expense of others.

At times, conclusions involve deliberate distortion; and on occasion can result in totally incorrect forecasts. A classic example of this is the predictions culled from surveys conducted during the election of Franklin Delano Roosevelt (FDR), who became the 32^nd president of the United States and the only US president to ever serve for more than eight years in office. A section of the US press, which was by and large against him and which played an important role in moulding public opinion, had pointed out that FDR was trailing in the contest, and yet, he won with a comfortable majority. In 1992, there was the famous example of a huge margin of error in the UK general election. Two exit polls predicted a hung parliament when, in fact, John Major maintained his position with a significantly reduced majority.

There have also been occasional examples of gross misreporting, as happened in the case of the presidential candidate Harry S Truman. According to the November 3, 1948 issue of the Chicago Tribune, Truman had lost the presidential race to the Republican governor of New York, Thomas E Dewey. The pronouncement achieved a measure of notoriety for inaccurate reporting. After he was elected, Truman flashed the newspaper to a cheering public at a stop at St Louis railway station. It was a moment that he relished. On the very same date, The Journal of Commerce went as far as to print eight articles, which focused on what the public expected of Dewey. Closer to home, there was the questionable research conducted during the last election where one local organisation predicted a close contest between the PML-N and the PPP, while another forecast pointed to a neck-and-neck challenge between Nawaz Sharif and Imran Khan.

At times, forecasters have enlisted the cooperation of the famous parrot, while in India, the faithful have been relying on the predictions of the astrologers.



Now for a bit of trivia and frivolity. Addicts with computers surf the net for lists of the 10 best of this or that. In my case, it is symphony conductors, classic films, Renaissance art and holiday resorts east of Myanmar. I avoid popularity lists in which researchers manage to elicit astonishingly revealing confidences, while imparting a sort of editorial semaphore of irony and disbelief. A case in point is the one which was reported in The Express Tribune in its January 3 issue, which carried the startling headline “Ash voted 4^th most beautiful woman in the world with Monica Belluci in the first position”. Since the popularity contest conducted by Hollywood Buzz is totally subjective and arbitrary, and based on pure choice, I don’t think one should take the matter too seriously. But it did raise a couple of queries in my mind. Why do people who organise and take part in these frivolous contests assume that one has to be an actress to be considered beautiful? And surely, in a country with a population of well over a billion people, there must be at least a hundred women who are not actresses, models or debutantes who are better looking than Aishwarya Rai.

Published in The Express Tribune, February 9^th, 2014.

Like Opinion & Editorial on Facebook, follow @ETOpEd on Twitter to receive all updates on all our daily pieces.