Monthly Archives: June 2016

Common sampling policy

I have the Wikipedia page which collates polling data on the EU referendum bookmarked, as all nerds do, and I was taking my daily cursory look at it this morning when something struck me about it. I fed the data into Excel to check I wasn’t overinterpreting, and there it was again.

The following graph shows sample size (on the X axis) plotted against Remain’s lead in the polls (on the Y axis).

graph 1

In general, the smaller the sample size, the larger the Remain lead is. This is largely because of the telephone polls: ORB’s phone polls always have a sample size of 800 exactly, and all the sub-1,000 sample polls are conducted by phone.

What to make of this? Here’s another chart: this time, assessing the the general election polls and the Conservative lead:

graph 2

It’s perhaps a little hard to make out, but if I’m not mistaken, the general election polls don’t show such a sharp disparity. Small sample sizes show exaggerated leads in one direction or another, but there’s almost as many figures below the X axis as above it. At the general election, the polls all clustered around a tie, which turned out to be very wrong indeed. Contrast that to the EU referendum polls, where there’s a much more marked profusion of big Remain leads in the little polls and a much more close picture in the big ones.

I’d hazard a guess that the bigger sample sizes are probably more accurate, but that depends on the polling companies having sorted out their methodologies. As it is, we’re flying blind.

Leave a comment

Filed under Politics

Some thoughts on status quo bias

Peter Kellner observed in the New Statesman yesterday that the last few referendums in the UK have all seen a late shift to the status quo, with the somewhat sui generis exception of the 1997 Scottish devolution referendum. He concludes from this assessment that history favours Remain this month, arguing that six of the seven referendums he discusses show support for the status quo.

Mr Kellner’s article has superficial appeal, but he’s guilty of leaving out almost half of the eleven referendums which have been held in the UK since 1975, which might undermine his thesis somewhat. In the below table I have listed all the referendums in the UK since 1975 (deliberately excluding the 1973 Northern Ireland border poll, which was boycotted by the nationalist community); the rows in red were excluded from Mr Kellner’s article.


*The 1979 Scottish devolution referendum resulted in a narrow (51.6-48.3) win for the pro-devolution side, but turnout failed to reach the 40% threshold imposed by Parliament in order for the referendum to have effect.
†The 1997 Scottish devolution referendum was made up of two questions, the first on the establishment of a Scottish Assembly and the second on whether it should have tax-varying powers. Both proposals passed.

By my reckoning, far from the status quo being the dominant response of the British electorates to a referendum question, it is actually slightly less likely than a vote for change – five questions asked were answered with the status quo response, while seven were endorsements of the proposed reform. Even excluding the 1979 Scottish referendum because of its turnout threshold and the 1998 referendum on the Good Friday Agreement because of its unique context brings us to an even total of five for five, with the status quo only taking a narrow lead if the 1997 Scotland referendum is collapsed into one. Mr Kellner has – perhaps deliberately – massaged his figures to portray votes in favour of change as significantly less common than they are.

Even his rather specious argument that the changes which succeed are those which endorse a ‘national consensus’  has virtually no value whatsoever. Perhaps the Scottish Assembly (as it then was) was indeed established as part of a national consensus. The same could not be said of the Welsh assembly powers referendum in 2011, nor even the 1979 referendum in Scotland which he explains away due to the insertion of a wrecking amendment by a Labour MP. It should be noted further that those two referendums exhibited the narrow margins which seem increasingly likely in the referendum this month, which should give both sides this time pause.

Ultimately, talk about bias towards the status quo misses something which may be quite important. In almost all previous referendums, the status quo is something which has existed for a very long time – since time immemorial, even. The Union between Scotland and England dates to 1707, and the Union of Crowns is a century older than that. Parliament has been sovereign in the North East, London, Northern Ireland, Wales and Scotland since the concept was first described. The first-past-the-post system has been the process for all British general elections since the expansion of the franchise, and predates even that for a much more limited electorate.

The exception, of course, is the 1975 European Communities referendum, where a relatively new ‘status quo’ – more or less a government policy – was endorsed merely two years after it was enacted. In 2016, the idea that the EU benefits from status quo bias and incumbency is dependent upon the idea that the EU has weaved into national consciousness as much as first-past-the-post, the sovereign Parliament or the very Unitedness of the Kingdom.

I’m not sure that’s so. Stronger In has been trying to run this sort of campaign, but it’s stuttering somewhat. My suspicion is that this is because the EU is almost a victim of its own success: so effective at inveigling itself into national economies that nobody even notices it’s there any more. Even whinging about Europe in tabloid newspapers has a detached quality, a sense of newness and alienness – consider the Metric Martyrs case for an example of this in action. It seems that the pro-EU side, which planned to run a simple defence of the status quo with the expected 2-1 walloping, is having to make the case for the EU afresh, as if we were being asked to join it rather than remain in it. Whether this affects the result remains to be seen, but it seems implausible to suggest the British electorate is on autopilot this time around.

‡Sort of. See MacCormick v Lord Advocate [1953] SC 396.

Leave a comment

Filed under Politics