While voting is not completed, it seems to me the informal vote is quite high. Is this right? If so, does it reflect a "plague on both your houses" attitude. Who lost votes to the "informals" if I am right.

Graham - There is nothing mysterious about qualitative versus quantitative polling, and I am no raw prawn. Space does not permit, but there is a lot more that could be said about relying on 136 self-selected respondents and one small focus group to make categorical judgments about why people voted in WA (as distinct from making guesses about a binary result). Perhaps Peter van Onselen might like to comment from an academic perspective.

I do not accept that your survey results show that media bias and electoral malapportionment can be "eliminated", any more than old growth forests for example, and we will have to agree to disagree on this. I maintain that political opinion polling is at a very superficial level in this country and deserves better methodology, including post-election polling. My guess is that the latter does occur at some level by the parties but is not released for public consumption.

I am interested in your comment that Textor-Crosby are being held responsible for the indecision/canal fiasco. Textor is widely credited as the originator of "push-polling" in this country, which says a lot about his methodology as a pollster. As for Crosby, if the WA result is anything to go by, his efforts in the UK might just backfire. And I find it hard to take NCB seriously. His integrity has been savagely attacked on Crikey as you would know, and this somewhat undermined his commentary.

Having said all that, this has been an interesting discussion, and I have learnt something more about Textor-Crosby. I hope the Liberal Party does too.

Grace, I think your confirmatory bias is showing. Academics do studies on even fewer samples. For example Barbara Pocock's recent report "Only a casual… How casual work affects employees, households and communities in Australia" was based on a sample of 55 self-selecting unionists. That wasn't raised at the time by anyone as a reason for dismissing her conclusions.

You're dead wrong about the state of opinion polling in this country. It is as good as anywhere else in the world, and in fact our On Line Opinion research is leading edge. We don't do exit polls here, but that is because they are basically a waste of money. News organisations, which do them in other countries often do so because the count is so excruciatingly slow. Here you know on the night. Political parties don't do it because come the election they've researched the electorate within in inch of its life and there's nothing more useful for them to find out. And they aren't necessarily very accurate - look at the last lot from the US.

You're very quick to jump on Crosby Textor. I said "I'd be surprised if they would have made a mistake like [the canal]." Today's newspapers tend to confirm that it was a couple of federal ministers who came up with the idea on the spur of the moment. Tex was probably pulling what little hair he has left out of his head at the time.

You're also wrong about Textor and push polling, which is not to say that no-one has push polled in Australia, but that the Canberra alleged "push poll" was nothing of the sort.

For the record Mark Textor is the best professional political pollster in the country, and Howard's election wins are a demonstration of it.

Of course I'd accept your dismissal of Crichton-Brown. If it's in Crikey it must be true! No need to deal with the substance of what he says. I'm not an NCB fan, but that doesn't mean he can't make sense.

Graham - it's too easy to throw around allegations of bias in this discussion. I am not and never have been a member of a political party, but I note you are the former vice-president and campaign chairman of the Queensland Liberal Party.

You have persisted in confounding my argument about opinion polling to find out WHY people vote, as distinct from which party is likely to win. Your 136 self-selecting respondents were asked which party they were likely to support, and provided a pretty good result indicator. However, you do not know any more than I do, why they voted this way, and one small focus group will not give you those answers with any confidence.

I have suggested that parties should do (if they don't already) some post-polling surveys to find out the WHY. You have confused this with exit polling to find out which way people voted, particularly in the USA which has voluntary voting. Exit polling makes little sense in this country as we both know because results are provided so quickly on the night.

You might believe that your opinion polling "is as good as anywhere in the world", by which you probably mean the USA, but I do not agree with you.

As for Textor and push polling you are either very naive or deliberately misleading. I invite readers to google "textor push polling", or go straight to Ken Parish's article on your own website, concerning push polling at the 1994 NT election, at http://www.onlineopinion.com.au/view.asp?article=1790, or to the comprehensive ABC Background Briefing on push polling and Mark Textor at http://www.abc.net.au/rn/talks/bbing/stories/s19393.htm.

Do you deny that following Liberal Party push polling in the 1995 ACT by-election, Mark Textor and Andrew Robb settled a defamation suit from the ALP candidate Sue Robinson for somewhere around $8000? Have you read any of the submissions or transcripts from the 1995 Joint Standing Committee on Electoral Matters (lapsed) inquiry into push polling?

Finally, it was you who first made positive reference to Crikey in relation to NCB, so your comment about my alleged pro-Crikey bias is a bit rich!

Grace, you say it is too easy to throw around allegations of bias, and then you accuse me of being biased. That's a bit rich isn't it? In case you aren't aware the term "confirmatory bias" refers to the tendency that we all have, me included, to see what we want to see.

You say "Your 136 self-selecting respondents were asked which party they were likely to support, and provided a pretty good result indicator. However, you do not know any more than I do, why they voted this way, and one small focus group will not give you those answers with any confidence." This suggests that you either didn't read our report or just didn't understand it. The 136 didn't predict the numerical result of the election, and the principal point of the survey is to ask people why they are doing what they are doing, not to count how many of them are doing it! Our quantitative measuring supports the qualitative.

You also seem to have a little trouble with the issue of push polling. None of the material you provide gives any evidence that Mark Textor personally push polls, although it does prove that Gary Morgan and Charles Porter conduct them. According to Ken Parish in his article, so does the ALP.

And I didn't accuse you of a pro-Crikey bias, I've just noticed that you aren't interested in exploring the actual facts and frequently seem to rely on someone else's opinion as absolute proof. So, instead of advancing an argument on its merits you cite some source and say, in effect, because it has been published it must all be true! I was being ironic.

Graham - it is really pointless arguing with you about who is biased, so I will not waste any more of my 350 words. I repeat my opinion that your polling does not explain with any confidence why people voted the way they did, and my disappointment that political opinion polling in this country is so limited, superficial, and methodologically weak. It has been reduced to a marketing exercise, not much better than branding shampoos, and USA-syle push polling, introduced to Australia in the early nineties by Mark Textor, the Liberal Party's "push polling guru", is just one sad exemplar.

I note that you have not denied the push polling defamation payout by Textor and Robb, and instead throw in a red herring about Morgan and Porter conducting the poll. They did not pay out because they were just the functionaries, acting under instructions from Textor and Robb, as we both know. Readers can make up their own minds on the widely published evidence about Mark Textor and push polling.

Finally, since we are both quoting Crikey favourably, here is an recent extract from Christian Kerr: "...The Libs had their special ops boys on the ground and in position [for the WA election]. And Colin Barnett had Team Blue...A hush-hush-HUSH kitchen cabinet of experts...The Libs heavy brigade were all deployed - from Mark Textor down - but Colin Barnett chose to listen to Team Blue...They had a finger in the Grand Canal...but who were Team Blue?"

It would seem that Textor was on deck for the WA election, but he was not responsible for the Grand Canal, as you suggested earlier. What was he doing then?