UK Aggregate Turnout is Mismeasured

Abstract

Many countries including the UK claim to measure turnout as a percentage of registered voters. However, what they actually report is the number of votes cast divided by the number of entries on the electoral register. Inaccurate entries and duplicates mean that the number of register entries almost always exceeds the number of registered voters. We show that official UK turnout underestimates turnout as a percentage of registered voters by up to 14 percentage points. However, national official turnout is actually somewhat closer to the UK-resident voting eligible population (VEP) turnout, with official overestimating VEP turnout by up to 3 percentage points or underestimating it by up to 5 percentage points when using turnout. However, the errors are considerably larger when looking at the constituency level where official turnout can overestimate VEP turnout by as much as 14 percentage points or underestimate it by as much as 9 percentage points. We develop an approach for combining information from multiple official sources to create estimates of UK domestic and overseas VEP turnout and RV turnout and provide these estimates for elections between 1979 and 2019.

Keywords: turnout, vep, vap, idea, aggregate, comparative politics, error, measurement, bias