Hacker News Reader

Comment by khzw8yyy | Hacker News Reader

Like many things that are orthodox in education research, this is likely wrong.

It would be nice for us (the masses) to believe that it is likely wrong, so we can sleep easier knowing that 'everything balances out'.

However perhaps this is an uncomfortable truth, that we may or may not choose to face.

yes!

take a sip at what educators are trying to push nowadays (not the [majority of] business paraphernalia from the ed. tech world)... it's usually about teaching kids types of skills not found at the grade; like emotional intelligence, resilience and even meta-analysis of their way of studying!

I'm not sure if I'm understanding you correctly - do you feel like emotional intelligence and resilience are not appropriate to teach kids? There are a number of fairly simple social and emotional tools/approaches I have learned in adulthood that I really wish would have been taught to me and those around me growing up.

How could your statement possibly be true? This is hugely correlative.

"Good at one thing" in children generally implies "parents are involved" which generally implies "better than average socioeconomic status" which generally implies "children at least relatively decent at most educational things". This is going to be doubly true for things like art or music (or anything with private coaching) which require a significant monetary outlay in order to be good.

About the only place where this doesn't hold are the extremes where high spectrum people can function spectacularly in math and science and yet be ferociously terrible outside of that.

> "parents are involved"

The most fundamental way in which parents are involved in the life of their children is through their gametes. Unsurprisingly smart kids almost always have smart parents, which is a double advantage.

It might be the most fundamental but it's unlikely to be the most important. That award probably goes to all the things parents don't do, like drinking during pregnancy or feeding their toddler nothing but cereal and pizza for months at a time.

I hate this kind of unfounded anti-scientism. If it's so obviously wrong, it should be simple to design a study that demonstrates the opposite, why don't you try?

I'm not anti-science, I'm anti performative science for the sake of "doing science". And anti "we can't do proper science so let's throw up our hands". Helping individual kids and yes, relying on anecdotes would do much more good in this specific scenario than a study.

Let me explain. Doing a "simple" study like this is nigh-impossible within the science establishment we have today. Like you said, let me try. Let's walk through the steps.

A PI would have to get a grant to help not the underserved/underprivileged kids, but rather those who are perceived to have an advantage. They have to work with hundreds of schools (gifted kids are rare). Within that set of hundreds of schools they have to control for cultural and socioeconomic backgrounds and so on. Having succeeded they'll have to get this published (and probably fight some accusations of being elitist in the process). All I am saying is: this PI better already have tenure.

This problem is not just about gifted kids. A commonly heard complaint in the field of psychiatry for example is that we only study the minds with pathologies. We don't at all look at how to improve things for healthy people. Come to think of it, that is true in medicine in general.

A well defined experiment on human Children is incredibly difficult to get past the IRB. Very hard to get informed consent from children and parents.

You are left with comparisons between school districts and/or states

Yes this is true. I got into education research from physics because I thought the dynamics of learning could be very interesting to model. But we simply lack the institutional ability to implement such observations necessary to build such a model. So I zoomed out and built models on how students made choices between semesters at their university (what class they will take, will they switch their major, will they drop out). But ultimately I got bored. Now I work on ai related to acoustics. It's a shame. There are data sets out there that can be mined, I was offered a postdoc with a company in Minnesota doing it with data from millions of school children. But they were more interested in me modelling whether a student would have a behaviour problem tomorrow so they could integrate it into their platform and sell as a product to schools. I wasn't much interested in helping to build something like that.

Reply on news.ycombinator.com