harriet Harman and You Gov

What was that all about last night in the BBC election night programme. Harriet Harman seemed to deride YouGov polling on the grounds that their polling had her placed last in the Labour deputy leadership elections and she actually won. Later on there was a communication from YouGov saying this was completely wrong and Harriet Harman humphed and said she'd have to check.

Have you checked your figures yet Harriet Harman?

I have done some digging and I cannot see on what you base your claim. Labour Home reported on 27 November 2006 that

"A new YouGov poll of nearly 2000 voters shows Harriet Harman is the voter’s choice for the next Deputy Leader of the Labour party.

...

Twenty nine per cent of swing voters said that electing Harriet would make them more likely to vote Labour, putting Harriet top and 10 points clear of Hilary Benn in second place. When asked who they would like to see as deputy alongside Gordon Brown, Harriet had the greatest appeal, beating not only Hilary Benn, but also Hazel Blears, Alan Johnson, Peter Hain, and John Cruddas.

Crucially for Labour MPs in crucial marginal seats, Harriet’s lead was stronger amongst two key groups: swing voters and women voters. "

Of course there was some spin i that report so let's look at what UK POlling Report said on the same day:

"The Indy reports a YouGov poll on the Labour Deputy leadership campaign. Since it comes via the Harriet Harman camp, the report predictably paints a very positive picture for Harman. But, I hear the cynics amongst you cry, what do the actual figures say?

YouGov asked two sets of questions about the leadership contenders. The first listed the potential candidates and asked respondents to say how they rated them on a scale of 0 to 10, with 0 meaning respondents felt “very cold & unfavourable” toward them, and 10 meaning they felt “very warm & favourable”. There was also the option of saying they had never heard of the person.

On this first question Harriet Harman was easily the best known of the candidates listed. Only 15% said they hadn’t heard of her, compared to 28% for Hain, 31% for Benn, 32% for Johnson, 38% for Blears and 58% for Cruddas.

This, of course, could also explain why Harman had the highest positive ratings. The figures quoted in the Independent are those for “swing voters” (defined in the normal way, as those people who say there is a fair chance they will change their mind before the next election). The percentage of people who have a positive opinion of Harriet Harman (giving her a score of 6/10 or more) is 24%, followed by Benn on 17%, Johnson on 15%, Hain on 13%, Blears on 7% and Cruddas on 5%. Amongst the public as a whole Hilary Benn is equal with Harman on 17%, followed by Hain 13%, Johnson 13%, Blears on 6% and Cruddas on 4%.

If you take into account net scores, subtracting those who express a negative opinion, then Hilary Benn receives the highest scores with -10 amongst swing voters, then come Cruddas and Harman on -14 (though Harman obviously has a far higher recognition rate than Cruddas), Johnson on -16, Hain on -23 and Blears on -27. Harriet Harman is, interestingly enough, the only candidate whose figures contrast strongly between the public and a whole and swing voters. Her net rating is -23 amongst voters as a whole, but -14 amongst swing voters.

The second set of questions asked how people would vote if each candidate were deputy leader. The question was presented in the terms of a ticket - i.e. leadership teams of “Gordon Brown & Hilary Benn”, “Gordon Brown & Hazel Blears”, etc, etc, so the actual change from now can’t be put down entirely to the deputy leaders, part of it must also be due to Gordon Brown. There are some interesting contrasts between the performances of the different deputy leadership candidates.

I am always wary to some extent of questions that ask people to say if X would make them more or less likely to vote for party Y. I’ve always suspected that some people say X would make them more likely to vote Y when they would actually vote Y anyway, and vice-versa. YouGov normally include, as they have in this instance, the option of saying “No difference - I would vote Labour anyway” and “No difference - I would not vote Labour anyway”, but I still have my doubts. I think people use such questions to indicate if they like a candidate or not. In reality, to what extent does the deputy leader of a party really impact voting decisions?

That aside, looking at the figures 15% of people say that they would be more likely to vote Labour if Gordon Brown was leader and Harriet Harman was deputy. However, another 15% of people say they would be less likely to vote Labour, so the net effect is neutral. This actually compares well to nearly all of the other candidates - Hilary Benn would also have a neutral effect, but everyone else would repel more people than they would attract. Hain’s net score is -6, Blears and Cruddas -8 and Johnson -9. Amongst swing voters Harman does even better, with a positive net score of +5, compared to -4 for Benn, -12 for Blears, -13 for Hain, -17 for Johnson and -19 for Cruddas (though given his low recognition, Cruddas’s figures on this question must be of very dubious worth).

So, while the figures reported in the Indy cherry pick the nicest findings for Harriet Harman, in actual fact the figures as a whole are pretty good for her too. Along with Hilary Benn she is one of the candidates with the most positive perceptions amongst the public, and amongst swing voters she seems to be the most likely to be an electoral asset. The poll also shows that she has the highest recognition rate (substantially higher than the serving cabinet ministers in the survey) which on one hand is a good thing in itself, but on the downside could mean that the other positive findings are all just a result of higher name recognition."

Finally I found this BBC page which actually reports that:

""I want to make sure we have the best leadership team that is best placed to win a fourth term," she tells the BBC News website.

And that, according to YouGov research commissioned by Ms Harman herself, would be a combination of Harriet Harman and Gordon Brown.

Of all the possible permutations of leader and deputy leader, the Harman/Brown dream ticket is the one with most appeal to "swing voters," the poll suggests, with 15% of those surveyed saying Ms Harman would make them more likely to vote Labour, compared to 12% for nearest rival Hilary Benn.

"I will be making that information available to colleagues, so that if for them the most important thing is getting a fourth term then I will be the best candidate," says Ms Harman. "

So what's the truth? Did YouGov ever show Harriet Harman in last place as Harriet Harman claimed last night? Was the YouGov poll actually commissioned by Harriet Harman and used to further her campaign? All very interesting, albeit mainly to the political anoraks like myself.