Comments on: KLR VU Guelph By-Election Poll

By: Scott's DiaTribes

Scott's DiaTribes — Mon, 25 Aug 2008 12:46:21 +0000

KLRvu polling methodology update…

Well, we’ve had quite a few of blogposts on the KLRvu pollsters and their polling numbers for Guelph throughout the blogosphere (including yours truly), and even one by David Akin looking at them. All this publicity no doubt thrills Allan Bruinoo…

By: Grant

Grant — Sun, 24 Aug 2008 20:59:43 +0000

Seems to me the big problem, amongst a number, is that they did not rotate the questions. That for sure invalidates results. As well a push button poll is far more suspect than what you indicate. The hang up rate is much higher and the number of calls they claimed to make to get that sample is indicative of a high hang up rate, much higher than other real polling firms. Add in the comment by the ‘pollster’ that undecided (in a by-election of all things) is insignificant tells me it might be more telling to whip spagetti at the wall and read the splash pattern for a prediction of the results.
My visit to the riding last week on business in two different ends of the city revealed very few Green and Conservative signs on private property. If signs mean anything, then these results contradict the in the ground action.

By: bigcitylib

bigcitylib — Sun, 24 Aug 2008 19:39:59 +0000

DL,

I think in the Morgentaler case you might find that significantly more Morgentaler supporters would hang up on a poll that characterizes him in this fashion.

By: Cap'n Crunk

Cap'n Crunk — Sun, 24 Aug 2008 17:24:15 +0000

Hmmm, another source that thinks they can report on poll results without even knowing the substance of the questions. How very professional of you.

By: DL

DL — Sun, 24 Aug 2008 16:25:25 +0000

There are some valid points – I’m more concerned that the only previous poll by this outfit was asking people if they thought “abortionist Henry Morgentaler” deserved the order of Canada – which is a pretty suspect way to word things.

We also have one previous example of byelection polls being done using this technique the results in the March byelections (the results of which the Green Party went to great lengths to try to publicize) – and in every riding the results were spectacularly wrong (ie: they said the Green party would come in second with 17% in that riding in northern Saskatchewan – they wound up a distant fourth with 3%!) – honestly a child playing pin the tail on the donkey could have done better!