3 ms·

I love how every reporter writing about this study implies that these gaps are somehow related to income inequality and health care. It makes so much sense, rig

I love how every reporter writing about this study implies that these gaps are somehow related to income inequality and health care. It makes so much sense, right? In fact, the study directly contradicts this. See figure 8 - both income inequality and health care were not shown to have a significant effect. https://jama.jamanetwork.com/article.aspx?articleid=2513561 https://jama.jamanetwork.com/article.aspx?articleid=2513561

Retric 11y ago
This study does not show what you suggest. The problem is when you have several causes for the same effect some of all of them can fall below the noise floor. Being poor means slightly worse health care, worse food, increased risk of violence, etc etc. Trying to seperate each of them out in a statistically meaningful way is difficult and expencive.
- yummyfajitas 11y ago
  Yes, the effects could be small rather than zero. Either way, why are reporters implying (while being careful not to explicitly state) that inequality and health care are significant? Essentially your complaint is that NHST could be more realistic and doesn't deal well with tiny effect sizes.
- llamataboot 11y ago
  One might even say, "separating them out" uses fancy regression models to obscure rather than explain the data. Being poor DOES mean all those things and likely health outcomes are overdetermined. We know that reducing inequality reduces all of these things - so let's just reduce inequality!
Tycho 11y ago
There is a staggering degree of beating-around-the-bush in these articles we've been seeing.
justaman 11y ago
Thats the name of the game now. Class warfare and racism get clicks. If there is no graph in an article about research it cant have my respect.
- minikites 11y ago
  Would you prefer USA Today, the "Cat In The Hat" of news sources because they sprinkle pictures everywhere so you don't have to read so many words?
- llamataboot 11y ago
  http://www.salon.com/2015/04/10/7_remarkably_cruel_instances_of_class_warfare_in_america_partner/ http://www.salon.com/2015/04/10/7_remarkably_cruel_instances...
NoGravitas 11y ago
I feel that you are misrepresenting the study here. The conclusions from the abstract: In the United States between 2001 and 2014, higher income was associated with greater longevity, and differences in life expectancy across income groups increased over time. However, the association between life expectancy and income varied substantially across areas; differences in longevity across income groups decreased in some areas and increased in others. The differences in life expectancy were correlated with health behaviors and local area characteristics. The figure you point to absolutely does not say that income inequality does not have a significant effect. Figure 8 shows that the association between life expectancy and income varied substantially across areas; differences in longevity across income groups decreased in some areas and increased in others. So yes, the differences in life expectancy are definitely related to income inequality. But the strength of that correlation varies depending on region.
- yummyfajitas 11y ago
  The pieces of the study you quote don't disagree with me. Read figure 8. The correlation between various local area characteristics (which include health care and income inequality) were correlated with bottom income quartile life expectancy. The confidence intervals for this do not exclude zero. What part of the study do you believe shows a statistically significant relationship between income inequality or health care and life expectancy?

Retric 11y ago

This study does not show what you suggest. The problem is when you have several causes for the same effect some of all of them can fall below the noise floor. Being poor means slightly worse health care, worse food, increased risk of violence, etc etc. Trying to seperate each of them out in a statistically meaningful way is difficult and expencive.

yummyfajitas 11y ago
Yes, the effects could be small rather than zero. Either way, why are reporters implying (while being careful not to explicitly state) that inequality and health care are significant? Essentially your complaint is that NHST could be more realistic and doesn't deal well with tiny effect sizes.
llamataboot 11y ago
One might even say, "separating them out" uses fancy regression models to obscure rather than explain the data. Being poor DOES mean all those things and likely health outcomes are overdetermined. We know that reducing inequality reduces all of these things - so let's just reduce inequality!

Tycho 11y ago

There is a staggering degree of beating-around-the-bush in these articles we've been seeing.

justaman 11y ago

Thats the name of the game now. Class warfare and racism get clicks. If there is no graph in an article about research it cant have my respect.

minikites 11y ago
Would you prefer USA Today, the "Cat In The Hat" of news sources because they sprinkle pictures everywhere so you don't have to read so many words?
llamataboot 11y ago
http://www.salon.com/2015/04/10/7_remarkably_cruel_instances_of_class_warfare_in_america_partner/ http://www.salon.com/2015/04/10/7_remarkably_cruel_instances...

NoGravitas 11y ago

I feel that you are misrepresenting the study here. The conclusions from the abstract: In the United States between 2001 and 2014, higher income was associated with greater longevity, and differences in life expectancy across income groups increased over time. However, the association between life expectancy and income varied substantially across areas; differences in longevity across income groups decreased in some areas and increased in others. The differences in life expectancy were correlated with health behaviors and local area characteristics. The figure you point to absolutely does not say that income inequality does not have a significant effect. Figure 8 shows that the association between life expectancy and income varied substantially across areas; differences in longevity across income groups decreased in some areas and increased in others. So yes, the differences in life expectancy are definitely related to income inequality. But the strength of that correlation varies depending on region.

yummyfajitas 11y ago
The pieces of the study you quote don't disagree with me. Read figure 8. The correlation between various local area characteristics (which include health care and income inequality) were correlated with bottom income quartile life expectancy. The confidence intervals for this do not exclude zero. What part of the study do you believe shows a statistically significant relationship between income inequality or health care and life expectancy?