Well by 'evidence' I mean, metrics that quantify the quality of the educational system. As I explained in another comment in this thread, I think a substantial

Well by 'evidence' I mean, metrics that quantify the quality of the educational system. As I explained in another comment in this thread, I think a substantial part of the goal of education, for many people, is in bettering their lives in the material sense; so, if we define 'quality' as 'the extent to which something fulfills its purpose', then the quality of education can, in some aspects, be measured by the amount that it allows people to better their lives; for example, to put that more concretely, in a country where a big enough portion of the people receives an education, and if we attribute an increase in GDP at least partially to an increase in education (a not unreasonable assumption, imo), and if other factors remain relatively stable, then we can measure the success of the educational system by using prosperity as a proxy for it. I proposed per capita GDP in the example above, there probably are others; maybe the weighted for cost of living median wage of those who actually got educations, for example. But analyses or data like that aren't that easy to come by, so if enough people get an education, we could use overall GDP as a proxy. I acknowledge fully that there are many more factors at play here, so I'm not proposing this as a one-size-fits-all measurement; but just as cash-on-cash-return can be a useful first-order indicator in some contexts and for some purposes to get an estimate of the performance of an investment; and just like lines of code can be a useful first-order indicator in some contexts and for some purposes to get an estimate of the complexity of a code base; GDP can be (I'm arguing) a useful first-order indicator of the extent to which an educational system fulfills its goals - which I'd argue is the 'quality' thereof. But, to answer your question more directly, I guess that any metric on the material (or maybe even mental) prosperity of an educated subset of a population would be 'good evidence'. It depends on what one defines the goals of education to be; but realistically, I think that the material part should be a sizable part of any such definition.