Search
Archives
September 03, 2006 - September 09, 2006
August 27, 2006 - September 02, 2006 August 20, 2006 - August 26, 2006 August 13, 2006 - August 19, 2006 May 09, 2004 - May 15, 2004 February 29, 2004 - March 06, 2004 February 15, 2004 - February 21, 2004 February 08, 2004 - February 14, 2004 February 01, 2004 - February 07, 2004 January 25, 2004 - January 31, 2004 January 18, 2004 - January 24, 2004 January 11, 2004 - January 17, 2004 Recent Entries
The day after
Evolution is the name of the game Before I go to bed... Just say "Yes" Beattie concedes nothing Springborg on message at last Cassie was right Live blogging from the tally room Seats to watch I see a bad moon rising Powered by
Movable Type 3.2 |
August 20, 2006Sunday Mail pollsTreat with caution. I wrote a column for this morning's Sunday Mail so perhaps I should be a little circumspect, but the polls that they published today have a limited ability to predict what is happening in the four seats covered: There were many problems with the polls. One was that they asked for generic voting intentions and listed all of the major parties. So, in Noosa one could choose to allocate a first preference to "Liberal" and "National", amongst others, even though the Nationals aren't running. You could also choose to vote "Independent". Now I don't know how many independents are running in this seat, but this is a very imprecise category at the best of times. Another problem is that they didn't use candidate names when asking for voting intention. In the case of this electorate the incumbent member is Cate Molloy who will be running as an Independent having been disendorsed by her party. It's likely that some people voting "ALP" will really be meaning that they are voting for Cate, as will a large number of those voting "Independent". So Molloy's vote is likely to be under-stated in some ways and over-stated in others. Further, it's not clear that respondents were even told what electorate they are living in. So they may well have answered "Liberal" or "National" because that is what they normally vote, when if they had known they were in Noosa and Cate Molloy was available, they may have voted for her. Of course the same could be true for the Liberal and Labor candidates as well. Another problem is that they have distributed preferences, but they do not appear to have allowed for people just voting one, as they are entitled to do in Queensland. That can have a significant impact on the vote after preferences. The figures that you can take notice of are Cate's approval rating (43 percent with 26 percent disapproving). Preferred Premier also needs to be treated with caution as it includes Bligh and Flegg as well as Beattie and Springborg. While we frequently ask this question in our "What the people want" surveys it will tend to understate the practical level of support for Beattie and Springborg by introducing another two names. There is no way of knowing whether those who support Flegg will also all support Springborg in a contest against Beattie, nor is there any way of knowing whether Bligh's support will all transfer to Beattie against Springborg. We ask the question because we are interested in how strong support for the challengers is and want a indicator of that. In an election campaign it is probably more pertinent to ask for approval and disapproval of each leader in turn, which is what we have done in our latest questionnaire, which you can fill-in by clicking here. Posted by Graham at August 20, 2006 10:38 PM |
Writers
Graham Young
John Black Mark Bahnisch Michael Lee Resources
|