The data behind the June general election

238

At the announcement of Theresa May’s snap election, a quick look at the polls would tell you the Conservative party is set for a sweeping win, increasing its majority in the House of Commons, and expanding May’s mandate for the Brexit negotiations. However, May’s decision isn’t as risk free as it looks. Taking into account the historical failure of UK polling, with a high margin of error, the Conservative’s 9 to 21 point lead isn’t as clear as it looks at first glance.

It could be argued that polls this far out aren’t accurate, but a look at historical polling tells us they aren’t that different to final week polls. In fact, looking at the polling averages from 1979 to 2016’s Brexit vote, the average error of the 50 day out polls has been just 6 points. Compare this to the final week poll average error of 5 points and you can see that while neither show particularly good polling the difference between 50 days out and the final polling is minimal.

Of course the stand out difference to this trend is Brexit, where polling missed by about 6 to 4 which by UK polling standards isn’t an outlier. Though public consensus and commentary talked up the probability of Remain winning, the data showed the race was close and well within the margin of error for Leave to win.

May’s Conservatives do have a substantial lead going into this election, with recent polls showing them 9 to 21 points ahead of Labour. Also, while the polls in the U.K. haven’t been very accurate, they’ve tended to underestimate Conservatives rather than Labour in the past, often attributed to what is often called the Shy Tory Factor.

If polls are missing election outcomes by 5 or 6 points on average, that means the margin of error (or 95 percent confidence interval) is very large indeed. Specifically, a 6-point average error in forecasting the final margin translates to a true margin of error of plus or minus 13 to 15 percentage points.

If the Conservative’s stay at their current polling average, their lead is more closer to an 8 point lead than a 21 point lead, much closer to a standard margin of error that could leave June 8th with more tension than expected with May’s announcement.