2 ms·

It'd be interesting to better understand the control for this research. The main premise here is that "those receiving pension benefits" (i.e. no longer workin

It'd be interesting to better understand the control for this research. The main premise here is that "those receiving pension benefits" (i.e. no longer working) have faster cognitive decline. But is there a correlation between the kind of work you did and receiving a pension? Perhaps the word "pension" means something a bit different here, but in U.S. English, we usually use it to mean a company-sponsored retirement plan, and that is something that is less common than it once was. Could it be that if you had a job where you knew you'd get a pension, maybe you didn't have to try very hard while you were working? That is, maybe you spent the last decade coasting and taking it easy before you retired. All of this could be invalid speculation if they just mean "retirement benefits" of any kind?

loeg 7y ago
Could also be confounding cause and effect (if not correlation and causation). Maybe people with medical/mental decline issues retire early at higher rates than background.
asdfasgasdgasdg 7y ago
Yes, it seems far more likely to me that those who start experiencing cognitive decline make the choice to take their pension benefits earlier. I'm really surprised to see so many people here commenting as if this study demonstrates anything.
waterhouse 7y ago
Yeah, that was my first question: Is this not just a hopelessly confounded correlational study? I was disappointed that the article didn't seem to talk about this. Well, at least it links to the paper, so here's a starting point: "In this paper, we estimate the causal effect of the NRPS program on cognition among individuals ages 60 and above. The expansion of the program affected an easily identifiable group, as the policy was introduced only in select areas. Our identification employs a triple difference (DDD) strategy. We exploit the staggered policy implementation between 2009 and 2013 and compare the cognitive outcomes of individuals 60 years and older who live in areas that implemented the NRPS program to the cognitive outcomes of individuals of the same age group who live in areas that did not implement the NRPS program."

loeg 7y ago

Could also be confounding cause and effect (if not correlation and causation). Maybe people with medical/mental decline issues retire early at higher rates than background.

asdfasgasdgasdg 7y ago

Yes, it seems far more likely to me that those who start experiencing cognitive decline make the choice to take their pension benefits earlier. I'm really surprised to see so many people here commenting as if this study demonstrates anything.

waterhouse 7y ago

Yeah, that was my first question: Is this not just a hopelessly confounded correlational study? I was disappointed that the article didn't seem to talk about this. Well, at least it links to the paper, so here's a starting point: "In this paper, we estimate the causal effect of the NRPS program on cognition among individuals ages 60 and above. The expansion of the program affected an easily identifiable group, as the policy was introduced only in select areas. Our identification employs a triple difference (DDD) strategy. We exploit the staggered policy implementation between 2009 and 2013 and compare the cognitive outcomes of individuals 60 years and older who live in areas that implemented the NRPS program to the cognitive outcomes of individuals of the same age group who live in areas that did not implement the NRPS program."