Notionally, we say that we would want to pay for universal education on the basis that eventually these children will be voting, and therefore in order to accept their verdicts, there should at least be the possibility that they are qualified to pass such judgment. Is that actually realized in our current system?
The measure of effectiveness to elections is rejecting the poor performers and secondarily the poor prospects; then choosing the best candidates amongst well qualified individuals. We consider that knowledge such as history, logic and reasoning capability to include statistical reasoning, and some bounding boxes on science and the like, are required for performing such selections.
Then we should be able to measure the effectiveness of education in turning out better voters, by an increase/decrease in decent voting decisions.
In the American Empire, clearly this is not happening at any level. 2016 was a disaster, but you could just as easily look at re-election of incumbents who clearly were not suited, and in general to the 20-year (starting in 2000) trend of not balancing the budget, winning the wars, and continuing not consistently to enforce the law, as the more complete statement of that inability.
We also should consider the longer history of universal education in the Empire as a comparison point. Clearly the election of candidates who wanted to continue segregation, female second-class citizenship, and other obviously wasteful policies, is a refutation that the universal public education typically provides a universal benefit, vs. an improved capability which does not have practical effect. We have to acknowledge limits to the conversion rate of the body of citizens.
If we consider the intermediate measures on what we consider the building blocks – the end results (like PARCC proficiency) aren’t great. Usually in the better school districts, 50% of the kids wind up making it to proficiency, so if our proposition about those building blocks is valid, at the maximum we could expect 50% accurate decisions + the dice roll on the other 50% of decisions. Even so, that’s quite a sizable percentage we would think would make decent decisions, and that isn’t what we see in the data. The poor performance in 2016 and the post 2000 (really starting in 1996, but close enough) era in the AE, indicates the elites who passed this competency test, still are failing at high rates.
In the present state, on the current curriculum and general approach, there clearly isn’t a case for spending money (let alone more money) on voting-specific capability. In the earlier grades, it doesn’t matter anyway, because the material is the same regardless of whether you talk about economic capability or voting capability. It’s when you get to high school, and to remedial activity related, that we have to ask this question, and hence, the current answer is, no, we don’t have sufficient evidence to spend money at a large scale for that purpose.