The Conservatives have outpolled all parties in local by-elections in Scotland over the past 2 years, they claimed. The party produced figures showing that since May 2003 it had polled 10,683 votes in local by-elections, compared with Labour's 10,219, the SNP's 9833 and hte Liberal democrats 73,05. The Tories with 23% of the share-of-the-vote tables.
Intriged I decided to investigate using this very usual list of by-election results from the Lib Dem website.
For a start there have been 34 local authority by-elections in Scotland since May 2003. Two of these North Morningside/Grange and Drumry cannot be counted in the same 2 year tally as the last five/six, so the Conservatives should really come clean and say the last 2 and a half years for a start.
Of these seats Labour had won 11 in May 2003, the Independents 10, the Tories 7, The SNP 3 and the Lib Dems 2. One ward Speyside has had two by elections the first being won by the SNP in August 2004. This represents a heavily skewed sample in May 2003 the Tories won 122 seats (10%), Labour 509 (42%), SNP 182(15%) Lib Dems 175(14%), Independents and others 236 (19%), compare that to the sample which had 21% defended by the Conservatives, 33% Labour, 9% SNP, 6% Lib Dem and 30% Others.
The next interesting statistic is that of the 7 Tory held seats which faced a by election 6 of these the Conservatives had over 40% of the vote in 2003, In fact on three occasions they had pulled more than 50% of the vote in 2003 and twice it was over 60%.
In 2003 in these 33 seats Labour polled 34.75%, SNP 24%, Lib dems 13.63%, Tories 24.3%, compared to 32.63%, 24.07%, 14.54% and 15.07% respectively. Showing a further Conservative advatage from the sample.
However, the change in share of vote in these elections makes interesting reading. Labour -7.8%, SNP +0.1%, Lib Dem +5.9%, Conservative +2% SSP -0.3%. So even from their strong starting part they were not hte biggest gainers.
However, best of all is that the fact that they have not been able to add up. Either since May 2003 or Jan 2004 I only get the Tories in 2nd place, which is exactly were they started in these seats in the first place. Labour 10,543 (May 2003 16,520), Conservatives 10,324 (11,551), SNP 9,434 (11,409), Lib Dems 7,644 (6,481) SSP 1,208 (1,577). Since January 2004 and therefore the last two years Labour 9,735 Conservatives 9,548, SNP 9,196, Lib Dems 6,746 and SSP 1,017.
Indeed the only party to actually increased its vote on about 3/4 of the vote in the by elections has been the Lib Dems.
So statistically this report was using false statistics, it is also using a heavily skewed sample in favour of one party over the others. I shall be writting to the Times seeking a correction.