AT: Inequality kills millions

[Chaos 2587]

There's a card that's been floating around lately, I don't remember the author's name, about how inequality kills people because there's a correlation between poverty and ill health. The bad identification strategy has always irked me, so here's a rebuttal: https://academic.oup.com/qje/article/131/2/687/2606947. Main limitation is, potentially, external validity, as well as the fact that it's difficult to summarize studies in round.

Hat tip: Via MarginalRevolution, this compilation of counterintuitive arguments about education and wealth: https://randomcriticalanalysis.wordpress.com/2016/05/09/my-response-to-the-nytimes-article-on-school-districts-test-scores-and-income/.

Edited March 31, 2018 by Chaos

[Miro 1470]

I think external validity loses this card the debate. 

 

Overall, our findings suggest that in affluent countries with extensive social safety nets, causal effects of wealth are not a major source of the wealth-mortality gradients

 

I think the article you posted in hat-tip encounters the same issues that other defenders of scientific racism encounter, namely the difficulty in distinguishing environmental and genetic factors.

Edited March 31, 2018 by Miro

[Chaos 2587]

I agree that external validity is potentially a problem, but the point is that the study raises the standards of argument. Anything better than a mere correlation is worth promoting. It also does a lot to counter the argument that the gap in outcomes is due to stress or status anxiety, which is something I've seen asserted frequently.

I'm not seeking to defend every idea in the compilation. I overlooked the section near the end, and I'm sorry about that. But previously, I had been under the impression that children of poor families attended schools that received less money per pupil on average, so the main argument was very surprising to me, and seemed worth sharing. I also thought it was insightful to test if these relationships existed in other countries, so we can rule out explanations that are unique to the United States. And I was surprised to encounter research suggesting that heritability is not depressed in low SES populations. On the whole, I think there's still a lot to consider in the link, even if many ideas are bunk.

Did you have a more specific complaint about how the author failed to distinguish between genetics and environment? Are you just saying that you think heritability metrics are inadequate for the author's point? I felt that showing gaps emerge in under 5 year olds strongly suggests an important role for genetics in explaining SES, with a high degree of relevance to many interventions falling under the topic. It's potentially a great argument that we should focus on figuring out how to do universal pre-K right, for example.

Edited March 31, 2018 by Chaos

[Jw4167 24]

bezrucka 14 is the card you are refering to

[Miro 1470]

Did you have a more specific complaint about how the author failed to distinguish between genetics and environment? Are you just saying that you think heritability metrics are inadequate for the author's point? I felt that showing gaps emerge in under 5 year olds strongly suggests an important role for genetics in explaining SES, with a high degree of relevance to many interventions falling under the topic. It's potentially a great argument that we should focus on figuring out how to do universal pre-K right, for example.

A bit swamped with psets right now, but I'll make the point against "showing gaps emerge in under 5 year olds strongly suggests an important role for genetics in explaining SES" that by age 3, there is a 30 million word gap between children from the wealthiest and poorest families. Obviously your point is still relevant to the education intervention, but I don't think it's sufficient to strongly suggest a genetic source of SES intelligence gaps.

 

Twin studies are sufficient to suggest a partial genetic source for intelligence though. 

Edited March 31, 2018 by Miro

[Chaos 2587]

That's fair, in light of findings like that I agree it's not necessarily sufficient to strongly suggest a genetic source. However, it is sufficient to rule out a wide swath of environmental explanations.

[Miro 1470]

Further, is the purpose of schooling to improve general intelligence or domain specific knowledge, the latter of which is arguably more important to social mobility.

[TheSnowball 1433]

bezrucka 14 is the card you are refering to

I'm not sure if that's it or not, but here's the Bezruchka card in case anyone's curious.

Bezruchka.docx