You include a copy of the Cognitive Reflection Test or similar in each ballot and weigh votes by the number of correct answers to the test.
This doesn’t quite solve the Schelling point problem. You start getting questions about why that particular test and not some other. You will also get problems related to Goodheart’s law.
You start getting questions about why that particular test and not some other.
Well… People might ask that about (say) university admission tests, and yet in practice very few do so with a straight face. (OTOH, more people consider voting a sacrosanct right than studying.)
ETA: now that I think about that, this might be way more problematic in a country less culturally homogeneous than mine—I’m now reminded of complaints in the US that the SAT is culturally biased.
You will also get problems related to Goodheart’s law.
Keeping the choice of questions secret until the election ought to mitigate that.
now that I think about that, this might be way more problematic in a country less culturally homogeneous than mine—I’m now reminded of complaints in the US that the SAT is culturally biased.
Also in the US the SAT is only one of the factors effecting admissions.
Keeping the choice of questions secret until the election ought to mitigate that.
Only partially. Also what about the people whose design the questions?
This doesn’t quite solve the Schelling point problem. You start getting questions about why that particular test and not some other. You will also get problems related to Goodheart’s law.
Well… People might ask that about (say) university admission tests, and yet in practice very few do so with a straight face. (OTOH, more people consider voting a sacrosanct right than studying.)
ETA: now that I think about that, this might be way more problematic in a country less culturally homogeneous than mine—I’m now reminded of complaints in the US that the SAT is culturally biased.
Keeping the choice of questions secret until the election ought to mitigate that.
Also in the US the SAT is only one of the factors effecting admissions.
Only partially. Also what about the people whose design the questions?
High-stakes testing, like the SAT, where voters - I mean, test-takers—have vastly more incentive to cheat, seem to do fine.
Come to think of it, the problem is that the people designing the SAT’s have fewer incentives to bias them then people designing the election tests.