Right. The point here is that whatever the marshmallow test does measure, it doesn't seem to measure what many economists would like to think it measures - at least, if the results of this paper hold. Noam, does that make sense?
-
-
Replying to @Noahpinion @tw_watts
Can you elaborate on the "what many economists think it means"? Cog ability has a known robust link to time preference (e.g. https://pdfs.semanticscholar.org/62b2/baac4399ea781e9a247affa8a56886e746f2.pdf … from
@GarettJones ), but why would economists think it measures behavioral problems?1 reply 4 retweets 12 likes -
Replying to @NoamJStein @tw_watts
Assuming this latest paper holds up, it means marshmallow test isn't a good test of time preference as a trait. Trait preferences (as opposed to state-contingent preferences) should be able to predict behavior in a way that's consistent across the lifespan.
2 replies 1 retweet 5 likes -
That doesn't mean that trait time-preference doesn't exist (though
@economistified) has done experiments showing it doesn't). Nor does it mean that the marshmallow test measures nothing of interest. It just means that the two aren't linked.1 reply 0 retweets 4 likes -
A big problem is that all this stuff is tied together at the genetic level. Check this out from Wertz et al: http://sci-hub.tw/http://journals.sagepub.com/doi/abs/10.1177/0956797617744542 …pic.twitter.com/jeBN9nGv57
1 reply 3 retweets 18 likes -
Replying to @karlbykarlsmith @NoamJStein and
Wait are you using twin studies for determining genetics in 2018? You can say heritable (which would include epigenetic, microbiome, etc) but "home environment" is genetic?
2 replies 0 retweets 0 likes -
Replying to @lcdriammdmph @karlbykarlsmith and
Everything is heritable, home environment too. Using it and assuming causality as most people do is obviously a problem. http://onlinelibrary.wiley.com/doi/10.1111/j.1601-183X.2009.00554.x/abstract … https://www.nature.com/articles/mp20152 … There's a meta-analysis with average h² of ~25% for home environment measures, but can't find it.
1 reply 0 retweets 2 likes -
Replying to @KirkegaardEmil @karlbykarlsmith and
Only 2-3% of total variance= "significant" but not impressive. But at least this used actual genetics. Too often people use twin studies that (at best) imply heritability but then use term "genetic". Missing h2 probably microbiome and or methylation IMHO
1 reply 0 retweets 0 likes -
Replying to @lcdriammdmph @KirkegaardEmil and
Furthermore the study linked to can't exclude reverse causation for the minor 3% variance. IE successful people more likely to mate, thus generating SNP pools that don't have functional importance.
1 reply 0 retweets 0 likes
I'd say almost no one in the field thinks h² primarily reflect microbiome or epigenetics. I don't know of any surveys, but the history of this field is that everybody else constantly brings up new ad hoc models to avoid genetic causation, and they never pan out. YMMV
-
-
Replying to @KirkegaardEmil @karlbykarlsmith and
No shock people that spent the last 40 years telling us that genetics explains everything only to have a rousing 3% of variance don't think that nongenetic modes of heritance matter. The only "avoiding" is the genetacists denying the clear reality of their failed GATTACA ideals.
1 reply 0 retweets 0 likes -
Replying to @lcdriammdmph @KirkegaardEmil and
And again, show causual alleles or accept reverse causation. Meanwhile, interventional studies using epigenetics and microbiome are yeilding results that knock out mice never did.
0 replies 0 retweets 0 likes
End of conversation
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.