…ople who are eligible to vote, in just the same way that the sample is smaller than the population. If voting were a statistical sampling exercise, then its validity would depend on the quality of the sample. A better constructed sample would trump the results of the vote. We don't do this. There is no sens…
This and the next line. Definitely not true. It’s the wrong way to look at it and it doesn’t trump the result in any event, it just trumps the certainty. I’ll come back to this (1).
The idea is to get as close as possible to the randomised (ensuring quality), double-blind controlled trial methods used in science, thereby reducing the effect covariates and confounding variables have on the result, which in turn, naturally deal with the quality of the sample issue in frequentist stats. All features of political voting in UK referendums cover the necessary characteristics:
1. Randomised — Ensures the quality of the sample. 34 million people voted in the EU Referendum. That is sufficient to ensure the randomised variability.
2. Blind — People voted using similar mechanisms used for general elections. i.e. votes were private.
3. Controlled — To prevent contamination, Electoral Commission ensure [UK] rules on free and fair elections are followed and crucially, guard against electoral fraud.
The only thing that doesn’t is the measure of significance.