I can't find it, but there was a great post on HN awhile back about how Finland doesn't really know what makes its schools so successful. Thus, articles like th

I can't find it, but there was a great post on HN awhile back about how Finland doesn't really know what makes its schools so successful. Thus, articles like this pull some facet out of the hat as the key differentiator. There was a good comment, suggesting that instead of modeling who has the highest test scores, instead model who is most successful at climbing the ranks of PISA. That's probably a better way of figuring out what contributes to success, because there are fewer independent variables.