How Votes are Collected and Counted
The manner in which votes are collected can heavily impact outcomes. For example: In serialized public voting, the first vote can distort the final result by as much as 34%.[[4]](#_ftn4) The question of how votes are collected has become increasingly important with the advent of the potential for e-governance.
There are only a few metrics that matter when thinking about this. None require that much thought:
- Is the voting public or private? Private voting is almost always better for the reason cited above.
- How hard is it to vote in an absolute context? While you could make a voting system that “filters” voters by formally making it extremely difficult to vote, we are not aware of any cases in which this has been done in practice (or at least not with any goal other than to bias a vote in favor of a specific party). In general, the current trend is to make voting easier (hence the rise of e-governances and mail-in ballots). We would say there is not an obviously right choice on this one but would still love to see some system at least experiment with artificially difficult voting (like a heated voting lever that hurts to pull) so that voting is intrinsically costly.
- Is voting equally difficult for all demographics? This could be a problem when looking at something like voting over the internet when not everyone has equal access to the internet and that inequality is expressed along economic or regional lines.
- Is the topic or individual in question bundled with other topics or people?
- Is it the case that you can only vote for one topic or person by also voting for another? This where riders come in: Agendas or candidates people would not otherwise vote for getting attached to an unrelated topic.
- Is the vote paired with a topic in a way that psychologically affects the vote? If you ask people to vote for multiple things at the same time, the way those things are paired can affect the aggregate vote outcome. For example, you might prime someone to consider the damage caused by muggings or vandalism right before having them vote on police funding.
Voting on a topic-by-topic basis exemplifies more nuanced complications around vote counting and collection. Topical voting is common with special ballot measures on issues, as seen in states like California where elections can more easily be held on specific measures. Despite superficially appearing to increase voter participation in a democracy, these systems are almost always a bad idea if the subjects being voted on cost additional money or need to be executed individually.
For example:
- If you allow votes to be taken on a case-by-case basis and ask a question like “Should teachers be paid more?” teachers will be far more likely to take the time to go out and vote on the measure even though the outcome affects everyone.
- “Should teachers be paid more?” intrinsically biases the voter towards yes, as the question is framed in the context of who gets more and not who gets less. Here it would make sense to frame said question against the cost: “Should teachers be paid more, using a tax increase of X%?”
- Even in this condition, it is easy to hide costs by indicating that costs will be borne by a minority e.g. “Should teachers be paid more, using an X% increase in taxes on companies in Y region earning over $Z annually?” In this case, the text is describing a cost that will hurt the region in terms of jobs, income, new companies, etc. but it would be difficult to explain those costs on a ballot.