Hacker News Reader

Comment by pastage | Hacker News Reader

Three times as likely to get college level education is amazing, children and parents really are the future.

Nurture is the major factor in nearly every aspect of human achievement. Alternative explanations justify genocide, neglect, and let adults off the hook.

> Alternative explanations justify genocide, neglect, and let adults off the hook.

While possibly true, none of those claims have any bearing on whether or not those alternate explanations are false.

Genocide? Umm, care to explain how that follows?

If some part of the gap in achievement is genetically determined (which by all measurements is the case.) than there is no way to close that gap. Children from poor parents stay poor, poor communities can not move up etc. What's proposed is that the only possible way eliminating differences is eliminating people (in either direction.). Of course this does not infact justify genocide or any of the sort, but that's the argument.

> Three times as likely to get college level education is amazing

To me this seems like a warning that the intervention is unlikely to have much effect as applied to society in general. Three times the likelihood of an impoverished Jamaican going to college isn't that impressive. On the other hand, if we accepted this assessment of impact in the terms being reported here, applying it to better-off people could lead to more than 240% of their children going to college!

...which again suggests that whatever's going on is probably better viewed in other terms.

I do not understand you at all, you comment can be read as a fairly common statement around here: education for rich kids is not always a good thing. While the article is about how families that have time to value their young are much more likely to educate them. How can you compare these two outliers?

You completely missed the point. OP is impugning the validity of one conclusion of the study - the claim "three times as likely to go to college". Such an outsized effect is only possible if we moved from an extremely minuscule likelihood to an only moderately minuscule. The claim is dubious when it comes from a sample of 105, a control of 65 and doubly so when the priors aren't mentioned. IE, would we expect .3 people out of 100 in this demographic to go to college?

I'm saying "increase the likelihood of the child going to college by a factor of three" doesn't make a lot of sense as an evaluation of the impact of this intervention, for two reasons:

- It's very easy to increase really small numbers by a factor of three, but this large percentage increase is not particularly meaningful.

- Many demographics go to college at such high rates that it would be impossible to increase those rates by a factor of three. Since it's impossible, the intervention probably can't accomplish it.

Nobody here tried to extrapolate the 3x into a general rule, except you. Not sure why you’re upset by something only you are doing.

Helping the less advantaged merits more of our attention than the well to do anyway.

> Nobody here tried to extrapolate the 3x into a general rule, except you.

The top-level poster replied to here clearly did.

Reply on news.ycombinator.com