Ask a silly question, get a silly answer (April 21 2005)
My colleague Paul Davies wrote a couple of days ago about these political surveys that purport to tell you how you should vote. Mary Ann Sieghart in today’s Times is similarly baffled by her experiences with these tests. In a spirit of scientific enquiry I find that I’m either a Green or a Conservative depending on which issues I’m thinking about on this survey; I’m a Liberal Democrat on this one’; and on this one my political position approximates that of the Dalai Lama!
The serious point, if there is one, is that it is very difficult to explain individual political choices even by such apparently rational criteria as the value statements these tests ask you to assess. The best explanation used to be inheritance – Butler and Stokes when they looked at this found that if both one’s parents voted for a party, you had something like an 80% chance of supporting that party too. It’s a bit more complicated now, and there are people whose position, like Paul’s, mine or Mary Ann Sieghart’s, that don’t map directly onto a party.
Ms Sieghart concludes her Times piece:
I can’t think how they came to their conclusions, but I shall have to ignore them anyway. My constituency is a marginal seat, which is a straight fight between Labour and the Tories. Voting for a minor party, even the Lib Dems, would be a pointless indulgence. I shall have to hold my nose and support one of the two main parties. But it looks as if no amount of ideological mapping will help.
But it is only the electoral system that forces her (and the rest of us) into such choices. In a PR system, perhaps a party could coalesce about Sieghart’s (perfectly consistent) centre-libertarian point of view. Perhaps under a multi-candidate system like STV the choice offered would be broad enough that one or more of the major parties would offer candidates that approximated her point of view. But even under STV I don’t expect the Dalai Lama to stand in my constituency any time soon – no system is that perfect, obviously.