Is a book by two UCLA sociologists, Vila Ortiz and Edward E Telles, published in 2008. It originated in a fair-sized data set (1576 people) collected in 1965, which was rediscovered in 1992. The original respondents and their adult children were interviewed. It shows quite clearly that although second-generation Mexican-Americans averaged more education and higher SES than the first generation, presumably because they knew English, there was no further improvement in the third and fourth generations. The gap remained substantial: the fourth generation had a college completion rate of 6%, compared to a rate of 35% for whites of that same era.
Which is pretty much what you see in New Mexico too, except that here we’re often talking about the fifth, sixth, and seventh generation living in the US
I don’t see much sign that the story is greatly different in Central and South America. Mestizos – whose ancestry is part Amerindian and part European (usually Spanish), make up most of the population in those countries. Their PISA scores are low – lower than those of Hispanics in the US. Performance in science and technology is more important than test performance – but Latin America’s low performance is consistent with their low test scores. This showed up in my Zones of Thought map.
Isn’t there reason to believe that this is all going to change radically for the better in the near future, powered by the strongest force in the Universe, wishful thinking? Nope.
Nobody knows how to substantially change this picture. Yet.