Public opinion polls, propaganda and Lansdowne

In the recent opinion poll on the fate of Lansdowne, the dubious stated assumption that the City owns Lansdowne may have unduly influenced responses, according to this writer.

Photo: Liz McKeen

By Randal Marlin

Public opinion polls serve the public interest in important ways. Those whose concern is to serve the public interest would like an accurate picture of what policy measures would best serve that interest. Knowing how people feel about the measures can be an important consideration for assessing their likely success.

Accurate polling costs money. If you are Conrad Black, you can get away with cost-free polling. Referring to his paper in Sept-Iles, he wrote with disarming frankness about his paper’s coverage during the 1973 Québec provincial election: “Our reporting was fairly balanced for most of the campaign, but we did an editorial sand-bag job on the P.Q., complete with publication of a poll indicating a Liberal victory. There was no indication of the number of people sampled so the fact that I consulted only seven people (including myself) never came to light.”

No reputable polling firm could get away with that completely unscientific, biased “polling.” They have to make use of acceptable science. But reputable polling costs a lot of money. Those who pay for polling may have an interest in the truth, but they may also have even more of an interest in truth varnished in a way favourable to their company, organization, party or whatever. The varnishing can result in misleading the public.

This brings us to the Ottawa Sports and Entertainment Group’s (OSEG’s) sponsorship of a poll about Lansdowne 2.0, the results and interpretation of which were publicized widely by CTV prior to the hotly contested vote to accept the proposed agreement between the City of Ottawa and OSEG.

The chosen pollster, EKOS, seems to have a good track record, and its science seems to be sound. One can learn a lot of truthful information on opinions about the Lansdowne 2.0 deal among people in the greater Ottawa region.

But with paid-for polls, one has to be very careful to check for biases favouring those who paid for the poll. In searching for bias, you need to know, for a start: 1. Who paid for it? 2. What were the questions asked? and 3. How were the results interpreted?

Against the background of the ambiguity in the concept of ownership, discussed in the August Glebe Report, I want to argue that two questions (there may be more) were biased and that the widely disseminated interpretation of the results was biased in favour of OSEG and against the City of Ottawa.

The first question I would impugn is Question 7. It begins, not with a question, but with a statement: “The arena and stadium at Lansdowne are owned by the City of Ottawa.”

This is a flat statement, one that doesn’t recognize that there could be other incidents of ownership conveyed to OSEG in the management, control and entitlement to income from the property. When you have a statement like that on something controversial, you are not just finding out information, you are seeking to influence people.

To see the relevance of this to the sponsor of the EKOS poll, we can turn to Question 14: “The estimated cost of Lansdowne 2.0 is approximately $400 million. Recognizing that Lansdowne is owned by the City of Ottawa, what proportion of the cost do you believe the City should be paying?” The question is followed by the possible choices: “0%, 25%, 50%, 75%, 100%, Don’t know/prefer not to answer.”

Doesn’t it seem obvious that someone who believes the arena and stadium are fully owned by the city will be more inclined to answer 100 per cent than one who believes some of the most important ownership rights are held by OSEG? Instilling the idea that the city is the only true owner carries with it financial implications that seem to favour OSEG.

Polls that seek to influence rather than find out truth about public opinion are sometimes called “push polls.” In the case of the EKOS poll, influence is very subtle, but it is there.

Interestingly, the poll does reveal something that is true and is the reverse of what EKOS considers to be true. This relates to the matter of EKOS’s interpretation.

The interpretation begins: “A clear plurality of the public believe that Lansdowne is owned by OSEG, only one in five recognize that Lansdowne is owned by the City of Ottawa. There is evidence that some of the concerns over the cost of Lansdowne 2.0 may be rooted in the incorrect assumption that the city is investing in privately-owned facilities.”

An alternative interpretation could be: “There is evidence that some of the concerns over the cost of Lansdowne are rooted in the correct assumption by a plurality of the public that some key ownership rights are transferred to OSEG.”

Randal Marlin is a long-time Glebe resident, former member of the city’s Lansdowne Park Advisory Committee and former president of the GCA, which actively opposed implementation of the proposed mega expansion of Lansdowne Park to Fifth Avenue in the 1970s. He is an adjunct professor in philosophy at Carleton University.

Pull-quote: “Polls that seek to influence rather than find out truth about public opinion are sometimes called “push polls.” In the case of the EKOS poll, influence is very subtle, but it is there.”