By arthur Weinreb
Tuesday, april 19, 2005Opinion polls that are commissioned by the media are different than those that are done for industry or private groups. When a poll is requested and paid for by a corporation or a political party, those organizations want to know the truth. The media on the other hand wants a story, regardless of the truthfulness of the responses that the pollster obtains. and when the media in question has a certain bias, they prefer to collect information that helps to bolster their particular point of view. By choosing who to pose questions to and by manipulating the language used in those questions, the pollster can help to ensure that the poll reflects the desired viewpoint.
Last week, Environics released a CBC-commissioned poll concerning views about a possible upcoming federal election.
according to Environics, 1,200 people over the age of 18 were polled. asking "people" about their views on an upcoming election is practically meaningless. These "people" could include for example, a 65-year-old who hasn’t voted since Daffy Dief was battling Looney Lester, a 45-year-old who has never voted and a 40-year-old who has voted in every election since reaching the age of eligibility. If someone has never voted or has not voted in the last few elections, they are not likely to vote in the next one. as interesting as their opinions might be, they tell us nothing about what will happen if an election is called. The only fair way to conduct a poll about an election is to poll those who are "likely voters"; for example those who have voted in at least the last three federal elections.
The poll asked respondents what the most important problem facing the country is. The top three answers were:
Healthcare 21%
Poor government/leadership 14%
Sponsorship scandal/Gomery Commission 10%
By choosing these categories, Environics helped the CBC (and, of course their political masters in Ottawa) obtain the desired results; that the most pressing problem is healthcare and not scandal. But the second and third categories overlap and many who think that the sponsorship scandal is the most important problem could have put their answer under poor government/leadership. By constructing the categories in the way they did, the pollster attempted to minimize the responses to the category of sponsorship scandal/Gomery Commission by giving respondents an alternative category. If you lump the second and third similar categories together, they exceed healthcare as the number one problem.
another question that the poll asked was, "are some political parties more ethical than others or do all political parties operate pretty much the same way?" This feeds into the belief that many people have that, "they are all the same" and it is hardly surprising that 68 per cent of respondents answered that all political parties are the same. The Canadian government is currently embroiled in a specific scandal, so a general question about ethics seems irrelevant to the current situation. Respondents should have been asked if they think that a Stephen Harper or a Jack Layton-led government would have been engaged in an adscam type diversion of funds. But if Harper and Layton would have been mentioned, it is doubtful that as many people would have answered that they and their parties would be as corrupt as the Liberals. But the greater number of people who say that they are all the same, the better the CBC likes it.
Readers and viewers should be very skeptical of polls that are commissioned by the media, especially those media outlets that have an agenda.