News for hackers. Stuff that matters.
Re:Randomized Control Trial of Pre-K on Children’s Skills Th... by ghaff on Tuesday October 27, 2015 @12:01PM (#10458870)
This result, among others, was discussed in this post (http://www.bloombergview.com/articles/2015-10-26/early-child...) and the writer's conclusions make sense to me. Small difference + statistical noise + conflicting study results could just mean that you're seeing no meaningful effect. There's probably no need to construct a twisted rationale over why this sort of program would actually produce net bad results.
Re:Randomized Control Trial of Pre-K on Children’s Skills Th... by slipjack on Tuesday October 27, 2015 @12:15PM (#10458976)
What's interesting is that the author talks only about impact on children. I was under the impression that a large part of the argument for universal preK (and universal day-care, like Head Start) was that it evens out how easy it is for parents to work when they have small kids. If that's the case, then it's not a huge problem that the academic results aren't great over time for the kids themselves if it means that overall family well-being is improved.

As a side note, Head Start is also interesting in that it, when it's done using blended classrooms, it also lets kids from diverse background interact with each other before they've learned a lot of prejudicial behaviors.

Re:Randomized Control Trial of Pre-K on Children’s Skills Th... by ghaff on Tuesday October 27, 2015 @12:39PM (#10459198)
It may well be true that pitching universal preK as universal child-care would be more honest. I suspect the problem is that it would also be significantly more controversial as it wouldn't really be "for the children and the future" any longer but rather, for better or worse, a straight income transfer to parents of young children.
Re:Randomized Control Trial of Pre-K on Children’s Skills Th... by vasilipupkin on Tuesday October 27, 2015 @04:18PM (#10460803)
yes. If you actually look at the results table in the study, the results are not even significant at 5% level.