Imagine the reaction on the part of Mayor Trent if, at some time in the future, a PQ government were to proceed with separation based on a questionnaire in which 22% of those eligible to respond indicated a wish to separate (of the 27% who answered, 87% chose this option). The rationale would be that the result could be extrapolated to the remaining 73% non-responders, even though they had expressed no such support. Mayor Trent (like many others) would be justifiably outraged.
Why then does the city persist in claiming overwhelming backing for the arena plan when in fact there is no basis, statistical or intuitive, to do so. The Mayor now routinely refers to the “83% of Westmounters who approved the project” (Westmount Examiner, Nov 04, 2010). Certainly, of the 27% of households who responded, 83% generally supported the proposal. What can we infer statistically, from the 3 in 4 households who did not respond - absolutely nothing. There wasn’t even the option to tick ‘undecided’.
We are all speculating when we try to extend approval, disapproval or indifference to those non-responders. The problem lies in the nature of the opinion-seeking process. The questionnaire was not a true poll – mainly because it lacked the element of randomness, was not a scientific sampling of the population (respondents self-selected) and cannot be considered professional. In fact it is probably the case that every hockey supporter in the city responded and that the remaining 73% have no interest in hockey or a new arena whatsoever – again, speculation. The informal poll conducted by Messrs. Hooja and Bossen, which demonstrated quite different results, at least fulfilled the requirement for randomness.
The questionnaire used in this case, really a tool to elicit support rather than a poll, is only convincing if there is an overwhelming response, of the order of 70% or more. Small populations with idiosyncratic features, in this instance hard-core hockey fans, introduce sampling bias due to self-selection. As a consequence, the results cannot be generalized to the balance of the population.
If the city truly wanted to respond to community needs it would not have obsessively pursued the 2-rink option to the exclusion of everything else. -
Troubling also was the inference that it was this plan or nothing. Some who agreed were afraid that if they rejected it, the city might return to the previous one which encroached on the tennis courts. Why the fear of alternatives? A more straightforward approach would have been to ask respondents to choose among various options. We now know that despite the claims of city hall to the contrary, the agreement with the government allows changes to the plan as long as the granting agencies are informed. A combined indoor pool/hockey rink could be built with these funds. If the city truly wanted to respond to community needs it would not have obsessively pursued the 2-rink option to the exclusion of everything else.
It is worrisome if the polling firm hired by the city asserts, as claimed by the Mayor, that these results can be extrapolated: “a professional polling firm….. had no hesitation in saying that 83% of all Westmounters agreed with what we are going to do”. Moreover, it could extrapolate “with certainty” that 83% of Westmounters are behind the arena/pool project. With certainty? Not even a margin of error? Wow, statistics is a lot simpler than we thought!
The city’s web-site states: Council would like to proceed with this project, but only if a majority of taxpayers is behind it. An approval rate of 22% (27% x 83%) has now miraculously been transformed into a majority.
Mavis Young
Melville Avenue
