On that poll of Lib Dem members
Some of the internet got very excited over a LibDemVoice poll earlier this week showing 46% of Lib Dem members don't want Nick Clegg to stay on as party leader at the next election.
The question itself was rather more nuanced than some of the comment upon it suggested - it gave respondents options of Clegg staying for the election, stepping down just before the the election or stepping down sooner than that (and also separate opinions for stepping down as leader and deputy PM). Most of the 46% of Lib Dem members that wanted Clegg to go were happy for him to stay on for now - 32% of respondents wanted him to step down as party leader at some point, compared to only 14% who wanted him to step down in the next year. It suggests to me that this is more about Lib Dem members thinking Clegg is probably not the leader to get them votes at the next general election, rather than a sign of unhappiness or opposition to him per se.
While I'm here I should write quickly about how representative the polls on LibDemVoice are. Stephen Tall and Mark Pack don't make huge claims about representativeness and are always quick to stress that they can't claim they are representative. This is admirable, but is sadly not a carte blanche, as however much the person doing a poll hedges it with caveats and warnings these are rarely picked up by third parties who report a poll and are more interested in making it newsworthy than reporting it well.
That said, I think they are actually pretty worthwhile. They have the huge advantage of being able to actually check respondents against the Liberal Democrat member database so we can be certain that respondents actually are paid up Lib Dem members and not entryists, pissed off former members, other parties supporters causing trouble, etc. LDV also have access to some proper demographic data on the actual membership of the Lib Dem party, so while their sample is unrepresentative in some ways (it's too male for example), they know this and can test to see if it makes a difference. They have also compared it against some YouGov polling of Lib Dem members which had very similar results, and actual Lib Dem party ballots, which had excellent results in 2008 and rather ropey ones in 2010. Mark Pack has a good defence of them here.
Of course, there are caveats too. The danger for such polls is if they end up getting responses disproportionately from one wing of the party or another, from supporters or opponents of the leadership. I am not a Lib Dem activist so such things may be over my head, but from an outside perspective the LibDemVoice website doesn't seem to be pushing any particular agenda within the party that might skew the opinions of their readers or which party members take their polls. If reading LDV does influence their opinions though, it could obviously make respondents different to the wider Lib Dem party (for example, here Stephen suggests Nick Harvey's increase in approval ratings could be the effect of making regular posts on Lib Dem Voice, which would indeed be a skew... but not on a particularly important question!)
I do also worry about whether polls that are essentially recruited through online party-political websites or supporter networks get too many activists and not enough of the armchair members, or less political party members (not an oxymoron, but the type of party member who joins for family or social reasons, because their partner is a member or because they want to contribute to their local community through being a councillor and the party is really just the vehicle).
All that said, while they aren't perfect and Mark and Stephen never claim they are, I think they are a decent good straw in the wind and worth paying attention to, especially given the verification of whether respondents are party members.
