> The new paper shows that this is not the case. The Marshmallow test still is predictive, but that seems to be because being able to take it correlates with having affluent backgrounds, not with it representing an inherent character trait that predicts anything.
And Smaug123 referenced that by saying
> Assuming the trait is either heritable or culturally imbued (and intuitively it seems very likely that at least one of those is true), it makes sense that rich parents are more likely to have children who are either natively or culturally high in The Desire To Pass Tests
You might have misinterpreted the article in question, since you write:
> it follows the assumption that being good at the Marshmallow test represents inherent character traits that are predictive of later performance. Except that the whole point of the article is debunking this notion [...]
The article does no such thing. It only shows that controlled for socioeconomic situation, there is no correlation. That does not mean that there is no predictive power.
First of all, I do not understand why you would argue that I don't understand the article because I supposedly claim that the test has no predictive power, since I literally write "the Marshmallow test still is predictive".
Furthermore, Smaug123's explanation here is flawed:
> Assuming the trait is either heritable or culturally imbued (and intuitively it seems very likely that at least one of those is true), it makes sense that rich parents are more likely to have children who are either natively or culturally high in The Desire To Pass Tests
The reason it does not make sense in this context is the very thing you cite yourself: when controlled for socio-economic background, the predictive power disappears. If the predictive power of the Marshmallow test could be ascribed to character traits, controlling to socio-economic backgrounds should enhance the signal of the predictive power, not remove it altogether.
> If the predictive power of the Marshmallow test could be ascribed to character traits, controlling to socio-economic backgrounds should enhance the signal of the predictive power, not remove it altogether.
This might be the source of the misunderstanding: What you say would only be true if character traits and socioeconomic background were uncorrelated, which is the opposite of what smaug suggested.
If the relationship is `y=x₀+x₁+𝓝`, then x₀ is more strongly correlated to `y-x₁` than it is to `y` alone. This holds even if x₀ and x₁ are strongly correlated, as long as they're not 100% correlated.
The question isn't whether you can produce example models where "What [vanderZwan says] would only be true if character traits and socioeconomic background were uncorrelated", it's whether it holds in general.
And Smaug123 referenced that by saying
> Assuming the trait is either heritable or culturally imbued (and intuitively it seems very likely that at least one of those is true), it makes sense that rich parents are more likely to have children who are either natively or culturally high in The Desire To Pass Tests
You might have misinterpreted the article in question, since you write:
> it follows the assumption that being good at the Marshmallow test represents inherent character traits that are predictive of later performance. Except that the whole point of the article is debunking this notion [...]
The article does no such thing. It only shows that controlled for socioeconomic situation, there is no correlation. That does not mean that there is no predictive power.