Ocean Kinetics - The Engineering Experts

Letters / Election forecasting

Shetland, along with fellow island constituencies Orkney and Na h-Eileanan an Iar, often tends to be among the first constituencies to declare results in Scottish Parliament elections.

Though ballot boxes have to be brought together from across the isles, Shetland’s small electorate means that counting can usually be completed fairly quickly.

This year, as social distancing requirements at counting centres will stretch the count across two days, it is likely Shetland will provide results especially early.

As well as offer relief for anxious candidates, the Shetland result may be watched from across Scotland as an indication of how the election is going.  But how much can Shetland tell us about voting patterns in the rest of Scotland?

Looking at every major party swing on the constituency ballot between Scottish Parliament elections from 2003 to 2016, parties in Shetland followed the national trend ten times out of 16 – or 62.5 per cent of the time.

For example, in 2016, the SNP, Labour and Conservative candidates all lost or gained ground in Shetland in line with the national result; only the Liberal Democrats gained ground while the party lost votes nationally.

This suggests just over half the constituency swings in Shetland may be replicated across Scotland – a decent correlation, but not really enough to be predictive.

The Shetland constituency can sometimes produce unusual results that go against national trends, such as Tavish Scott’s landslide win in 2016 or Billy Fox’s strong 2011 performance as an independent candidate (though I suspect that will not be the case this time).

Fortunately, there are two ballots in this election. The regional list ballot allows voters to vote for parties to ensure more balanced and proportional representation in the Scottish Parliament.

With less scope for individual candidates to produce local deviations from national trends, Shetland’s list vote tends to be much more aligned with national movements.

Examining swings between elections for each of Scotland’s five main parties, parties rose and fell in Shetland in line with Scottish trends 17 times out of 20 – or 85% of the time.

If that tendency continues this year, four out of five party swings in Shetland will be replicated across the rest of Scotland.  That’s a pretty good indicator for how the rest of the election might go.

Though Shetland’s list vote tends to reflect the direction of travel between parties at a Scottish level, it has historically told us less about the size of the swing.

For example, while the SNP increased their Shetland list vote share in 2011 and the Conservatives in 2016, in line with national movements, on each occasion the size of their vote share gains was less than half that seen across the rest of Scotland.

Shetland is a distinct political community and often votes according to its own interests and patterns.  However, voting behaviour is influenced by national trends, and if Shetland does indeed provide a quick result next Thursday this can give us an early indication – but only an indication – of how the rest of the election may transpire.

Mathew Nicolson