Wednesday, October 11, 2006

Excuse Me While I Piss on this Poll

The PGIB sent me their Leadership Poll news release and results breakdown. This is where they are saying it is a three horse race from a “PIGB poll of 600 Alberta PC Party members…over a three week period in Calgary, Edmonton, Central and Southern Alberta.” The other fact they tout is this is a poll where PC Party members only were being polled. That is a good thing - but they did it so badly they destroy any data supported evidence for their conclusion. They may be right but this "poll" is no proof of it. Duh!

There are some serious credibility problems with these poll results, some of which have already been pointed out, like the randomness of the inputs with no northern Alberta input, how can you be conclusive? The other irony is they say in the news release their poll is “statistically significant at 19 times out of 20” but fail to mention the margin error. Duh!

Statistics, questionnaire design, data collection and results analysis is part art and part science but it is a professional activity that “ought not to be tried at home” by amateurs, especially amateurs with an agenda! People like the PGIB ought not to be doing this because it is misleading and pure political misdirection unless they are not serious about their credibility as an organization’s reputation involved in political lobbying – apparently one of their core activities. Duh!

I would be interested in seeing where they got the list of names came from to do the PGIB survey. If it was names only from their own membership list (which I suspect) that also taints the results, given that they are an Alliance based, social conservative organization that would not therefore reach the full range of Progressive Conservative members. It also excludes those Albertans who intend to become PC Party members to participate in the Leadership. These people are not represented in their data inputs either. Duh!

This is at best described as a crude political tactic or cheap trickery. It is not a poll. If it uses PGIB members who say they are also PC Party members but that is validated through the PC Party membership list the premise of their survey is even more unfounded. The Undecided and refused to answer numbers are very high too and are in themselves sufficient to challenge the conclusion they reached even if the methodology itself was not so amateurish...which it is. Duh!

Thirdly they asked the wrong question: “Are you currently a member of the Alberta Progressive Conservative Party?” That question is insufficient because it ignores those who intend to buy memberships between now and Dec 2 in the one person one vote system. And how come it took 3 weeks to gather the data? Do they not think opinions vary dramatically over such a long data collection period and the results are invalid as a result? Or am I giving this charade too much credit. Duh!

This is a very laughable and sad situation at the same time. Emmanuel Kant is giggling in his grave where people like Chandler, a poster boy for the far right, proving to be practically post modern personal examples out to prove that their political ends is justifiable using whatever means necessary - all in the name of winning and at all costs.

This is a dramatic example of the kind of society and polity we will have in Alberta if either Oberg or Morton gets to govern. What kind of Alberta will it be with these kind of people and organizations setting the agenda and providing the "values" perspectives about the future of Alberta.

Thx Craig Chandler and the PBIB for this “disclosure.” To the rest of us - Be Afraid! Be Very Afraid!