"This is a dangerous idea." Ideas are dangerous. I think you mean "this is a flawed article because it starts out to buttress a preconception rather than exami

"This is a dangerous idea." Ideas are dangerous. I think you mean "this is a flawed article because it starts out to buttress a preconception rather than examine the data and then find a conclusion." "we don't REALLY know how random this is" Seriously? It's probably at least as random as any random assignment to condition in US college labs (I speak from a great deal of experience here). We're talking huge n values and a system that presumably at least has some level of public scrutiny (again, unlike US college labs). Why compare to US college labs? Because that's where most accepted social science research labs comes from. In any event, I see two major flaws in the initially cited study that are far more obvious and problematic than this. 1. Not everyone is Korean. Culture does matter. (Not everyone is a US college freshman either, of course.) And worse: 2. The outcome is measured entirely in terms of going to a 4-year college (good) vs. not going (bad). Not, for example, being "happy" 20 years later. Or graduating from college even. Anyway, what would we expect from a rambling rant hosted on the domain singlesexschools.org.