Posted by Simon Hix and Nick Vivyan.
We have updated our election prediction model, based on national voting intention results from all polls by YouGov, Ipsos-MORI, Populus, ICM, ComRes and Angus Reid with fieldwork up to and including 23 March.
Here is a figure showing the time trends in our “pooling the polls” analysis. The dots in the figure show the results from the various polls and the shaded areas around the lines are the 95 per cent confidence intervals around the mean standings of the parties. Basically, since the beginning of the year the Conservatives have lost approximately 4 points while Labour and the Lib Dems have gained approximately 2 points each.
Our results suggest that as of 23 March the national standing of the parties was 36.5 per cent for the Conservatives, 30.5 per cent for Labour, and 20.2 per cent for the Lib Dems. What would this mean in terms of seats in the Commons?
The following table shows how these national vote shares might translate into seats under different assumptions: namely, a uniform change in party support across constituencies, compared to a model of how marginal seats in different regions in the country deviate from national-level changes in party support.
Hix/Vivyan Seat Projections for 25 March
| Hix-Vivyan pooling-the-polls model | Comparison with other predictions | |||
|---|---|---|---|---|
| Party | Assuming a uniform swing | Controlling for a regional marginal seat effect | Electoral Calculus Seat Projection (up to 18 March) | UK Polling Report Seat Projection (on 25 March) |
| Conservatives | 284 | 306 | 302 | 286 |
| Labour | 278 | 258 | 262 | 285 |
| Liberal Democrats | 55 | 54 | 53 | 48 |
| Other Parties | 33 | 32 | 33 | 32 |
| No. of seats Conservatives short of a majority | 42 | 20 | 24 | 40 |
In other words, even assuming that the Conservatives do better in marginal seats than nationally, it currently looks like we’re heading for a hung parliament.
