The BBC's Bill Bush answers your questions about the pollsters' performance...

Question from Helen Thomas, rogert@cix.compulink.co.uk

A question not about opinion polls but real polls.... What would have happened in Winchester if the result had been a dead heat instead of a 2 vote majority? Has a dead heat ever happened in a British general election?

Dead heats have happened in council elections. The law says that if there is a dead heat then the candidates concerned draw straws, or cut cards. However, before that takes place there is immense scrutiny of "spoilt papers" to see whether any of the strange markings that some people make when voting can be interpreted as a valid vote. There hasn't been a dead heat in Parliamentary elections on the modern franchise.

In the cotswold constituency the tory vote fell by about 5,000 and the Labour vote rose by 6,000. You report this as a 1.2% swing from LibDem to Tory. Would you like to try again? If you look at both the remnants of the old Cirencester & Tewkesbury constituency you might note that the voting patterns are very similar. Since there was no referendum party in Tewkesbury the clear message is that the Cotswold Ref's took votes from Labour and LibDem and left the tories untouched. Adding these votes back into the two parties in equal amounts makes the two results from the Constituencies remakably similar - and if you assume that the average ref voter thought the LibDems were softer than Labour on Europe and therefore more votes came from the Lib Dems than Labour the similarity becomes striking. The clear message from both Constituencies is that: 1. There was a massive swing from Tory to Labour, but that even with this huge swing the LibDems were not unseated from second place. 2. The referendum party did not take votes from the tories. 3. The tories for the fist time in recent years got less than 50% of the vote. Under PR they MIGHT not be in power. I do not understand how the BBC obtained the comment under either constituency's entry, and suggest you should review it. Did the result prove inconvenient for the preconceptions or is it that the computer doesn't deal with the complicated ones? One last word. During the campaign the national media left us strictly alone, mostly because we don't fit conveniently into the media regions and therefore don't pull in many viewers. The pundits all knew what the result was going to be here, anyway. For this I give you my heartfelt thanks!

Swing is a crude measure best suited for a two party system. However it is a simple device to explain the relative shift between any two parties in any two elections. So although the main voting shift was ultimately from Con to Lab, we can still show what the swing was for Con and LibDem. We decided that rather than show all the swings between all parties we would simply show the swing between the top two parties. I don't agree though that the move was mainly in terms of Con switching to Lab - for example, if 5,000 LDs switched to Lab, and 4,000 Tories switched to LD, then that would look like a mainly Con to Lab move in terms of the overall changes but the churning under the surface would in fact be quite different. Obviously many voters did switch direct from Con to Lab, but that is by no means the whole story

Dear Mr Bush, Just recently Ken Livingstone said that Tony Blair should raise taxes, Tony Blair is effectively blessed with the luxury of being able to ignore him and his views. Such a large majority allows him to almost "impose" legislation and force it through even though there might be a lack of support for it in his own party (though obviously he would not want this). I imagine also that it would be more difficult for the Labour whips to prevent dissention when Labour MPs know they can dissent without crippling their party. Will this mean that we will see a lot more criticism of the Labour party by these "rogue" MPs, while the Conservatives become more united having seen the damage done by disunity.

All new Governments enjoy a "honeymoon" period when party dissent is quiet and public abuse limited. However it always ends - in Mr Major's case with the ERM crash and devaluation in September 1992. Once events crowd in dissident voices begin to feel that they have the room for manoeuvre to begin to criticise. In the best interests of the party, full and free debate etc etc. For example there are several LabourEuro sceptics who will never vote for EMU. There are also many who have actively campaigned against privatisation and they made need some persuading if the new Government decides to continue the Tory programme. Dissent is always present in parties, the question always is how it is handled. The electorate doesn't mind a row, so long as there is a clear resolution of the problem. What they hate are debilitating, continuing splits. Labour's majority means that they afford a few rebels, but they will not want the numbers to be large, nor for the debate to be rancorous.

Did the boundary reforms cost the Lib Dems a relatively safe seat in Liverpool. I refer to the old seat of Mossley Hill, which now lies within Riverside and wavertree? OR would that seat have been taken in the Labour landslide. What career is David Alton now hoping to follow, following his standing down at this election?

The boundaries certainly dispersed LD votes and mixed them in with biggish Labour majorities. However the general results in urban areas in the north of England suggests that David alton would probably have lost even if he had stayed on and had fought on unchanged boundaries. After all, Liz Lynne failed to hold Rochdale, and Chris Davies also failed to hang on to the (altered) by-election seat in Oldham East. David Alton was given a peerage by John Major on 19 April. I'm sure he will use that as a platform for his particular interests, perhaps including his Pro Life views.

A lot of the talk at Conservative Campaign HQ as reported last night on the election special seemed to be suggesting that the polls were not matching up to the results of the Conservative party canvassers. Do you think that canvassing has outlived its usefulness, based on this failure?

Party workers are bound to say that their canvass returns look good. But if a canvass is carried out properly it is very helpful. It was canvass returns that told Edwina Currie that she had no chance in Derbyshire South, unfortunately she told the Express, who then ran it on the front page. That's why party workers usually fib in public. Labour by contrast were fibbing in the other direction: they didn't want complacency to set in, so they said that their their returns showed them only modestly ahead. Ultimately, the purpose of canvassing is to locate your vote so that you can get it out on the day; ie it isn't really the same as an opinion poll where you want to know what everyone thinks.

What would have been the outcome of the seats if done under proportional representation. If you are unable to tell me this where would I be able to find out. Thanks Yvonne

Depends a bit on the PR rules adopted. For example some so-called PR systems have a qualifying threshold - you have to get over 5% to get elected. Some PR systems are based on multi-member constituencies, so again the arithmetic means that very small parties find it impossible to get over the qualifying "quota" level.

Anyway, the vote shares were approx Lab 44%, Con 31%, LD 17. There are 659 seats in the Commons, of which 18 are in Northern Ireland. If there had been PR in the GB seats (641) then Labour would have won 285 seats, the LDs 110 and the Tories 201.

How much influence do you think the national press had in influencing the outcome of the election, particularly the defection of the so-called Tory papers? Do you think their reporting of the national opinion polls (and that of the broadcast media of course) makes people think "I want to be on the winning side"?

The newspapers are part of the climate of opinion. There is no evidence that newspaper endorsement influenced people much in the campaign - the damage was probably done in the preceding 5 years of abuse heaped on Mr Major and his colleagues. What's interesting is how similar the election result was to the local and European election results of the last four years, and the adjusted opinion polls. It looks as if the nation made its mind up sometime in 1992/1993 and didn't really shift thereafter.

CAN YOU PLEASE EXPLAIN WHY ALTHOUGH THE TORIES CLEARLY SUFFERED A SHARP FALL IN THEIR OVERALL MAJORITY IN 1992 THEY ARE SAID TO HAVE THE RECORD NUMBER OF VOTES FOR A BRITISH PARTY?

The Tories got a record vote in 1992 partly because the UK electorate gets bigger each year and there was very good turnout - probably because everyone thought that the election would be close and so it was especially worth voting. In 1997 the turnout fell quite sharply - on average by 6% or so. That means that Labour has won a huge majority but with fewer votes than the Tories got in 1992. However, we don't have compulsory voting in the UK and if people stay away then it leaves the decision to those who turn out. The Tory loss of seats was so dramatic because of the fall in their vote rather than the increase in Labour's. The LibDems and the SNP also did quite well in those seats where Labour was not in contention - so for example the Tories got far more votes than the LDs in Scotland and Wales but won no seats there at all.

Does the government have to abide by the result of a referendum. eg. on single currency ?

Depends on the precise legislation introducing the referendum - referendums can be advisory or mandatory, but in the UK the constitutional principle is that Parliament is sovereign ie it takes the final decision. However the EMU referendum would only take place once Parliament had made a recommendation, and so ratification after a "yes" vote would a formality