I have read the paper[1], and its discussion section supports the conclusions drawn by other readers, who are smarter than I am[2][3][4].
Here's one excerpt from the discussion:
> "When considering how our results might inform intervention development, recall that models with con-trols for concurrent measures of cognitive skills and behavior reduced the association between delay of gratification and age-15 achievement to nearly zero."
To put it another way, when controls for education and socioeconomic status are added, there is no significant effect.
You absolutely did not pay the attention while reading. At age 15 indeed the association diminishes to almost zero; the original experiment never claimed otherwise. The original claim was that association is strong at early childhood; nonetheless it is pretty stupid to believe that a marshmallow is a sufficient reward for 15 years old person.
Insulting me and assuming I didn't read my links doesn't make your argument stronger. It makes you sound insecure about your argument, which you should be, because you have missed basic facts about the study.
> At age 15 indeed the association diminishes to almost zero; the original experiment never claimed otherwise
The original experiment (and you, based on what you implied about people who fail to take Covid precautions) absolutely did claim an association at age 15 and beyond. That was the whole reason it became famous.
> nonetheless it is pretty stupid to believe that a marshmallow is a sufficient reward for 15 years old person
I can't tell if you're trolling or not, but I'll engage with you as though you're being sincere.
The data for 15 year olds is academic achievement, not a repeat of a marshmallow test.
From the study:
> "Academic achievement was measured using the Woodcock-Johnson Psycho-Educational Battery Revised (WJ-R) test (Woodcock, McGrew, & Mather, 2001), a commonly used measure of cognitive ability and achievement (e.g., Watts, Duncan, Siegler, & Davis-Kean, 2014). For math achievement at Grade 1 and age 15, we used the Applied Problems subtest, which measured chil-dren’s mathematical problem solving."
Here's one excerpt from the discussion:
> "When considering how our results might inform intervention development, recall that models with con-trols for concurrent measures of cognitive skills and behavior reduced the association between delay of gratification and age-15 achievement to nearly zero."
To put it another way, when controls for education and socioeconomic status are added, there is no significant effect.
1. https://sci-hub.se/10.1177/0956797618761661
2. https://jasoncollins.blog/2018/05/31/the-marshmallow-test-he...
3. https://www.theatlantic.com/family/archive/2018/06/marshmall...
4. https://www.vox.com/science-and-health/2018/6/6/17413000/mar...