I don’t personally think young people are bad MPs; that’s just what other people may think. The average voter, aged 50, probably has a tendency to think that someone aged 25 is too young, that’s all. On a personal note, I’m younger than any MP except Jo Swinson, and I’d like to see a lot more young MPs elected. It’s strange that the youngest male MP, Dan Rogerson, is nearly 10 years older than Pitt the Younger was when he became Prime Minister.

The Times Guide for 1987 gives the Staffs SE result as SDP 14,230 and Labour 13,230, so the last figures could have been erroneously duplicated or muddled before being written on the results sheet, and/or the first two fiugures could have been transposed. Bt it doesn’t seem to fit in with the theory that “it went against every trend that was taking place, locally or regionally” because the reported swings were SDP +6.0%, Labour -3.0% : slightly unusual but not a big enough discrepancy to arouse suspicion, and within a reasonable margin of error.

I am reminded of two of the earliest results on the BBC results programme from 1987:

(1) Cheltenham was erroneously reported as a Liberal gain, and was corrected a few minutes later. This was a straightforward cock-up which happened because (a) the returning officer went straight from the candidates’ names to the number of votes, without pausing and without naming the party; (b) the BBC commentator spoke the party names after the candidates’ names, thereby speaking over the first part of the number. Thus the BBC mis-heard and mis-typed the number.

(2) Sunderland South was reported as having a Conservative vote of 1,621 instead of 16,210. This was also corrected a few minutes later, and was explained as “a mistake by the returning officer” which presumably meant that the R.O. read out a number different from what was on the paper in front of him.

I still haven’t found a full explanation for this one either:

(3) Sedgefield by-election: after the declaration in Ealing Southall, where I was the OMRLP candidate, it occurred to me that I hadn’t heard the result from Sedgefield yet. I went to the media room and asked one of the journalists. She told me that my colleague Alan Hope had got 129 votes, but that it had earlier erroneously been announced as 147. (She didn’t specify exactly how the correction or discrepancy had been done). When I got home afterwards, I phoned Alan and he told me that he had got 129 votes. I also phoned Sedgefield council and they confirmed that it was 129. However I became aware that the R.O. had actually read out 147. I don’t know why, and as far as I know the result has still not been posted on the Sedgefield council website.

It has been suggested by some people that the vebal declaration by the R.O. is the “official” de jure result, although the Sunderland South example (above) would seem to contradict that. My logic tells me that the written published result should be regarded as the de-jure one, because it is prima-facie irrefutable, whereas there could often be doubt about exactly what was heard or said at the verbal declaration. (For example, at both of the parliamentary by-elections at which I was a candidate, the R.O. did not repeat any of the numbers even when they were partially drowned out by cheering.)

No way could there have been a seat with nothing particularly unusual about it (Bradford W 1997) showing a 1% fall in the Tory vote surrounded by seats with 7-10 increases.

I presume you mean “decreases”.
But I think you are wrong in saying that there was “nothing particularly unusual” about Bradford West in 1997: the two main candidates were from different ethnic minorities, and there may have been idiosyncratic side-effects from that fact, just as there was in Bethnal Green & Bow.

John - I thought he was citing Bradford West as a seat that did have something unusual to explain the difference.

I meant to say Bradford W WAS unusual (1997)
hence a result completely out of line,

Chertsey and Walton must have been a huge mistake.
It looks like it should have been perhaps 32,000
for Pattie, not 25,000.

John Cartwright:

The Times Guide in 1987 for Staffs SE actually made a mistake with the probable mistake, if that makes any sense: the reported result in the vast majority of cases (including the Times Guide 1992) was: SDP - 14,230, Labour - 13,874, not 13,230 (that is obviously a misprint of the SDP figure.)

But it was out of line, because virtually every surrounding seat in Staffordshire had Labour improving and the Alliance declining. For example, in the next door seat of Mid Staffordshire, Labour overtook the Alliance for 2nd place. In 1983, Labour was well clear of the Alliance in Staffs SE; I for one simply don’t believe that the Alliance could have overtaken Labour in that seat in 1987, especially since the Lib Dems couldn’t even poll 10% in 1992, and Tamworth - the main town in the seat - has always been very poor for the centrist parties. In council elections, the Lib Dems don’t bother to contest 9 out of 10 of the local wards. Obviously, they were a bit stronger in the 1980s, but not that much.

Two separate points.

Firstly, this is a key seat the Tories need to get a Westminster majority. The recent YouGov poll would seem to suggest that a win here is possible, but Populus indicates this seat is beyond the Conservatives at present.

I have looked at the comments for a number of seats the Tories need to win to get an overall majority, and they do not in general appear to be getting them at the time of writing. In general, they are not reaching the older English town seats with a substantial middle class component. I suspect that the Tories are instead piling up votes in their own seats and the first tier of marginals they need to deprive Labour of an overall majority. Is there any reason to think the Conservatives are currently doing better than their national average in Stockton?

One specific issue in the North East is Northern Rock: is this having any electoral impact in Stockton South, and if so what?

The second point concerns the accuracy of results, which has been discussed in relation to this seat. Various instances have been highlighted above about gross mistakes in reporting results, such as identifying the wrong figures to the wrong candidate. But is there any evidence about the accuracy of a traditionally conducted election count? Presumably it would be possible for some university researcher to produce statistics experimentally, by doing a number of dummy counts. Of course some evidence could also be produced from recounts.

In principle, presumably the accuracy of election results could be assessed in the same way as scientific (or social scientific) measurements. This should enable an estimate to be made of the standard errors associated with the figures in a declared result, and hence by extension whether the votes for two candidates (or between a candidate’s voted and the level need to save a deposit) are sufficiently far enough apart that the difference is unlikely to be due to chance errors in counting. If the difference is not significant, of course there should be a recount.

Clearly any such investigation, and subsequent procedures, would need to take into account methological consequnces of counting methods, e.g. a difference of 50 (or whatever) votes may be generated by misplacing a bundle of votes as they are stacked.

Obviously, if new methods of counting are introduced their accuracy needs to be compared by political scientists with previous methods to show that they are better (or if cheaper, at least not worse).

Another implication of research into such matters is that it should be possible to produce guidelines for returning officers and candidates as to when a recount is justified, although there would need to be room for discretion, e.g. if there is suspicion that a ballot box has not been fully emptied, or if there is likely confusion between candidates’ names.

Academics have researched most things these days, so they have presumably looked at these issues! If anybody knows of published information in this area, perhaps they would like to post details.

JJB, elections for the local authority are all out rather than rotated. There were elections in 2005 because of dramatic boundary changes and then in 2007 to bring things back into their usual cycle, the next set of Borough Council elections will now not take place until 2011.

Yarm’s population is reasonably stable, not much in it either way.

Thanks, we’re working hard!

I think the tories have set themselves backwards here since coming to power in the local authority with labour, however grudgingly.

48% increase in councillor allowances, voted through by tory councillors, is an absolute disgrace.

talk about setting yourself up for a fall.