Letters
Statistical reasoning
Sir: Ferdinand Mount's sensible discussion of the electoral impact of the Alliance (3 April) rightly stresses that this cannot be measured on the basis of 1979 voting which, in its absence, would certainly be different today. As he says, the Alliance is intercept- ing votes which, cast for the Conservatives in 1979, would now be falling to Labour if the third choice did not exist.
But he fails to apply his own sound reasoning to Hillhead, which in 1979 went against the trend. Nationwide the swing was 5 per cent to the Conservatives; in Warring- ton 4 per cent, in Croydon NW 3 per cent, in Crosby 6 per cent; in Hillhead 1 per cent to Labour. Mr Jenkins had no 'soft' Tory voters to intercept (and presumably no `soft' Labour voters either, who had con- sidered switching in 1979 but did not do so in the end). He thus had a much more uphill task than the candidates in previous by- elections, quite apart from the existence of the SNP as a rival attraction for the disillu- sioned. To compare the percentage changes since 1979 with the other by-elections therefore underestimates his performance and overestimates that of the Conser- vatives,
P. M. Williams
Nuffield College, Oxford
Previous page
