Tuesday, November 15, 2011

Survey's 'majority' doesn't add up

TUESDAY, 15 NOVEMBER 2011 02:01 LETTER TO THE EDITOR Re: "Municipal taxes too high: survey" (Lethbridge Herald, Nov. 8, page A1). It's interesting to see that 49.9 per cent is labelled "a majority." There's a common phrase: "There are lies, damned lies, and statistics." However, it's not the statistics - the numbers - that lie, it's the interpretation of them that can be misleading. The research quoted in the article found that 15.8 per cent of the respondents think their taxes are far too high and 34.1 per cent believe taxes are somewhat too high. Adding these two together, a common and legitimate practice in interpreting survey data, gives the 49.9 per cent figure. But 49.9 per cent is not a majority. Just ask the supporters of Quebec separation in 1995. But I digress. If adding response percentages from a survey is a legitimate exercise, let's add some other response percentages. The survey found an almost identical number, 46.4 per cent (almost within the margin of error, I might add), felt that property taxes are at about the right levels in relation to the services received from municipal governments. Also, 3.7 per cent think their taxes are too low. (This is made up by summing 3.4 per cent who think they are somewhat too low and 0.3 per cent who state that they are far too low. Adding these together leads to the conclusion that 3.7 per cent think taxes are too low. Thus we could conclude that 50.1 per cent (46.4 + 3.4 + 0.3) think taxes are about right or (even) too low. This figure, 50.1 per cent, is a majority, by the way. They'd likely object to a cut in services done to reduce taxes. As I said, it's all in the interpretation. When one of several possible interpretations leads to the conclusion that taxes should be lowered, and amplifies it by suggesting that 49.9 per cent is a majority, one wonders why. Just asking.
Mark Sandilands Lethbridge

No comments:

Post a Comment