Well, I’m not going to pretend that the Lib Dem Voice members’ surveys predicted the outcome of the party presidential race within a statistical margin of error. But it is interesting to see – in the first test of the surveys in an internal election – that they were reflective of how the wider membership actually voted.
October survey:
Chandila Fernando – 3%
Lembit Opik 10%
Ros Scott – 61%
I don’t know yet, but will definitely vote 23%
I won’t be voting – 3%
There is one measure on which the LDV members’ surveys are certainly unrepresentative of the membership (sadly): while in October only 4% of LDV-reading party members said you wouldn’t vote at all, in the actual ballot some 50% of the party membership chose not to.
Share:
4 Comments
“50% of the party membership chose not to [vote]”
Well, except that some of the non-voters will:
have moved house
have failed to receive the ballot papers
be working away/ on holiday
have died
Most organisations would be thrilled to have 50% of their members voting in an internal election.
I think the low turnout explains why the LDV poll was so close. In short, the activist base which dominate the poll also have a disproportionate influence in the ballot itself. If it had been a higher turnout, I suspect it would have been less accurate.
Out of interest, how many people voted in the poll? (Just wondering, so I can work out if that really is within a reasonable margin of error.)
Andy,
28,782…