How Satisfied Are You With This Poll?

One evening last week, the phone rang. At the other end was a polite man who said he worked for Ipsos MORI, the market research company. He asked me if I would mind answering a few questions. Having nothing else to do, I agreed. Besides, I always wondered who this mysterious ‘British public’ was that ended up being polled; and realising that on this occasion I was one of them made me think I might as well make use of the opportunity.

The polite man on the end of the phone started asking all kinds of questions about my view of the political landscape in the UK. I answered the questions as diligently and truthfully as I could.

Then I came to the biggest question of all, Question 10. Question 10 went like this: “Overall, how satisfied or dissatisfied are you with the way the government is dealing with immigration and asylum?” I had to say whether I was very dissatisfied, fairly dissatisfied, fairly satisfied and so on.

Faced with Question 10, I could only answer that I was very dissatisfied. Why? Well, because of several things. Because of the deliberate political elision between questions of immigration in general and questions of asylum. Because of the outsourcing of housing for asylum seekers to odious and incompetent private companies like G4S and Serco (see the article here) who skim off money for their shareholders whilst failing in their duty of care. Because of the brutality of an asylum system that allegedly awards immigration officials with holidays and shopping vouchers for winning tribunals.

For all these reasons, and a hundred more, I told polite man on the phone that I was very dissatisfied. I then waited for Question 11, which would allow me to contextualise this answer. But the interview moved on, and there was no chance to provide any context. By the time I got off the phone, I felt uneasy. I felt as if in answering this question transparently, my voice was being added to a pool of data that would suggest an entirely opposite view. I thought of the sketch from Yes Prime Minister about political polling.

That evening, I wrote on Facebook and said that “I came out of the interview with the horrible suspicion that my discontent will enter some mass of statistics that will be used to argue for some kind of insular, bigoted, small-minded, Faragism.” A friend suggested I get in touch with Ipsos MORI, so I sent them a tweet (because that’s what you do these days…) to say that I was concerned about the poll. To their credit, Ipsos MORI responded very swiftly to say that they were always concerned about transparency and balance, and to ask what I thought was wrong with the poll. I had a friendly exchange of emails with the Chief Exec, who conceded that there was an issue here. The issue is this: that one could be as far right as it is possible to go and think the policy too weak, or as far left as it is possible to go and think the policy too strong, and either way this would be lumped in as dissatisfaction. I replied to thank him for his response, and to say that I mainly worried about how this data was to be used politically. It is not that I think the poll is deliberately misleading; only that it has not taken account of the context in which it is set, and it has inherited a number of assumptions from that broader context that make it somewhat questionable.

So I was not entirely surprised when last night I saw in the Evening Standard a big picture of Nigel Farage, and beneath him a headline reading, “Six in 10 unhappy with Government on immigration, exclusive poll reveals” (I’m not sure why the Evening Standard need to use the adjective “exclusive”, but that’s beside the point). Despite the big mugshot of Farage, the article — by Joe Murphy, the Standard‘s political editor — was careful not to say directly what the nature of this dissatisfaction was. The editorial in the same paper was less careful, however, reading:

Some six out of 10 people are unhappy with the Government’s handling of the issue, including more than half of young people. Plainly the Government has failed to meet the Prime Minister’s commitment to reduce net immigration levels to the “tens of thousands” rather than the hundreds of thousands; the most recent net migration figure was 182,000.

Of course, Londoners also appreciate that migrants contribute enormously to the economy, especially the service economy and the health service, but many people are concerned about the impact of large numbers of arrivals on the provision of public services and housing. Any audit of the costs and benefits of migration must take both into account.

In other words, the implication is being made very strongly that the dissatisfaction with the government’s stance on immigration is that it is not hard-line enough, and that there are not enough controls upon immigration. This implication is itself not surprising. Given the current political (and media) climate, this was always going to be the assumption that this is what the data meant, which is why the question was leading question. And yet, strictly speaking, this conclusion is not warranted by the data at all. It might be that most people are dissatisfied because they perceive immigration levels as being too high. That would not be surprising. But “might is not good enough. In terms of Question 10, we cannot reach any clear conclusion about the reasons for dissatisfaction with the government’s stance on immigration. In other words, if it is a question that a) tells us nothing concrete, and b) is liable to be over-interpreted to fit political agendas, it is a question better unasked; or else it is a question that, if asked, should be asked alongside further questions to provide more context.

This small insight into the way that political facts are conjured up out of over-interpreted data, and then used to feed into wider agendas, is something that is both instructive and dispiriting. Even if Ipsos MORI are careful not to over-interpret their data themselves, the construction of the poll is troubling because questions about satisfaction and dissatisfaction about such politically charged issues — without being contextualised by questions about in what way and why — can lead to poor understandings, poor reasoning, poor judgements, and ultimately poor political and ethical decisions.

So how satisfied was I with this poll, where one is very satisfied, two is satisfied, three is neither satisfied nor dissatisfied, four is dissatisfied and five is very dissatisfied? I would have to say five.

Leave a comment

name*

email* (not published)

website

Follow

Get every new post delivered to your Inbox

Join other followers: