Those Oldham East and Saddleworth polls: what do they mean?

by Stephen Tall on January 9, 2011

We reported last night the results of the three opinion polls published ahead of the Oldham East and Saddleworth by-election this Thursday. Let’s take a look now at the only two questions that matter when reading these polls…

1. Will the opinion polls prove accurate?

Opinion polls in by-elections have a notoriously patchy record — unsurprisingly, as the pollsters are unable to refine their techniques over time and measure their snapshots against actual results as they are able to with their general election predictions. For example, pre-by-election polls in Glasgow East (2008) and Blaenau Gwent (2006) showed the wrong winners.

They are also one-off surveys, and the pollsters can quite reasonably point out that their findings were accurate at the time, but that subsequent campaigning and/or the ‘get out the vote’ operation affected the final result.

That we have three opinion polls — two of them by long-established pollsters, ICM and Populus — all pointing to the same order of the parties (Labour first, Lib Dems second, Tories trailing) suggests we can have some confidence that the current order of the parties is accurate.

However, one of the reasons pollsters in previous by-elections have found it hard to assess the situation is the difficulty of predicting turn-out (usually far lower in by-elections) compared to general elections (where you can assume c.65% turn-out).

By-election success is dependent on two factors united by a common theme: which party can best motivate (i) its activists, and (ii) the voters — because it’s the work of the activists on the ground who will determine how many of the party’s identified voters are knocked-up on polling day.

There are two encouraging aspects of the by-election campaign for the Lib Dems.

First, that the party has had no difficulty in persuading activists from all over the country to travel to Oldham East and Saddleworth, with an estimated 250 out in force yesterday. No matter that many loyal party members are still smarting from the party leadership’s U-turn on tuition fees, they still want the party to win. Such enthusiasm is going to be crucial if the party is to buck the national polls at the English local, and Scottish and Welsh national, elections in May.

The other encouraging, closely related, aspect is this: the party still knows how to mount an effective local campaign. According to Lord Ashcroft’s poll for the Sunday Telegraph, reported here at ConservativeHome, voters are well aware of the the Lib Dems’ campaign: in 6 out of the 7 measures of contact with voters (leaflets, posters, phone, etc) it is the Lib Dems which have the best contact rate. Only on actual door-knocking are Labour in the lead, and then by just 2%.

2. Will any of it matter?

If the polls do prove accurate, Labour are heading for a win, with the Lib Dems second and Tories in third. And if that’s the case, it’s hard to see Oldham East and Saddleworth being a by-election that’s long remembered — after all, how often do governing parties defeat opposition parties in by-elections?

There are two-and-a-half ways in which it might prove a ‘game-changer’.

The first is if the Lib Dems defy the polls and grab a spectacular victory (still possible).

The second is if the Tories defy the polls and relegate the Lib Dems to third place (extremely unlikely) — given the Tories appear to have tried their best not to split the anti-Labour vote, it would’ve been highly embarrassing for the Lib Dems not have been at least runners-up.

And the half is this: if the scale of Labour’s victory were to be off the Richter scale — ie, a swing from the Lib Dems to Labour of 25-30% or more, the kinds of swing we’ve seen governing parties suffer in the past. The ICM and Populus polls indicate a swing away from the Lib Dems of c.8-10%. Would Nick Clegg and the party be happy with that? Of course not. But compared to what commentators have been forecasting is happening to the Lib Dem vote, there would be some measure of relief, too.

Want to prove the opinion pollsters wrong? There’s still time! The last week of the campaign will prove crucial — click here to do your bit to help the Lib Dems’ campaign.