Well, "large scale"... it's a culturally/racially homogeneous country of a little over 300k people. Whatever works there will probably not work anywhere else
I've enjoyed many of your comments, but these kind of phrases are almost impossible to argue with. Why won't they work? What would it take to make them work? Moreover, if it is true that heterogeneity implies data-driven impotence, how do we exert change to govern ourselves in a smaller scale, more homogeneous way?
The solution, as you just said, is to govern ourselves in a smaller scale. Apply data-driven solutions (like this one from Iceland) to smaller communities, one by one. Also, make sure those who apply these solutions are knowledgeable about how those smaller communities work.
Funnily enough, TFA says precisely this.
The not-so-well-behaved usually aren't square, and usually don't fit in anywhere.
I also somewhat disagree with the notion that systems working in smaller countries cannot work in larger or more diverse countries (esp. regarding free education the US). It may be really, really damn hard, but such is the price of advancement of society.
Just a few numbers to put in perspective:
* Iceland's population (as of 2013) was 323,000.
* As of 2015, the US had 58 cities whose population was over 320,000. Combined, these cities account for about 16% of the total population of the US.
Though, even that's a bit misleading, because city populations aren't actually as meaningful as metropolitan populations (since city boundaries are arbitrary, and traveling between a city and the adjacent suburbs is easy). Since Iceland is an island, it's more elucidating to compare it to metropolitan areas:
* The US has 155 metropolitan areas whose population is larger than the 320,000. Together, these cities account for 74.7% of the US population. So, three-quarters of people living in the United States live in a metropolitan area whose population is greater than that of the entire country of Iceland.