3 ms·

The article features no actual math to calculate the magnitude of the effect, even though such data are available. I would guess it would be a couple of percent

The article features no actual math to calculate the magnitude of the effect, even though such data are available. I would guess it would be a couple of percentage points shift at best.

mxkopy 3y ago
From the study: > We analyze this mechanism applying propensity scores matching and multivariate regressions on data from MIDUS I (Midlife in the United States: A National Study of Health and Well-being) and its 10-year mortality follow-up. Results show that health differences between 10-year survivors and non-survivors explain 56% of their differences in socio-political participation. https://www.sciencedirect.com/science/article/abs/pii/S0277953618300108 https://www.sciencedirect.com/science/article/abs/pii/S02779...
gerdesj 3y ago
The article is peppered with links to studies and quotes. for example: "as a 2017 CDC study showed" - showed is a hyperlink to: https://edition.cnn.com/2017/05/02/health/african-american-death-rates-cdc-study/index.html https://edition.cnn.com/2017/05/02/health/african-american-d... which links to https://www.cdc.gov/vitalsigns/aahealth/index.html https://www.cdc.gov/vitalsigns/aahealth/index.html The article quotes: "For blacks 18 to 64, the data showed that they were at a higher risk of early death than whites." I would not expect a journo piece to feature actual maths but I do want to see sources and we have them. I have no idea what your final sentence really means but it does: "guess". Sources?
nathan_compton 3y ago
The way our first past the post system works, a few percentage points is enormously impactful. A few percentage point swing in Florida could change presidential elections permanently.
storkhm 3y ago
I would think cigarette smoking is a big variable in the effect too. People living in poverty, uninsured, didn't graduate high school are all variables we know correlate with a higher % of cigarette smoking.

The article features no actual math to calculate the magnitude of the effect, even though such data are available. I would guess it would be a couple of percentage points shift at best.

mxkopy 3y ago

From the study: > We analyze this mechanism applying propensity scores matching and multivariate regressions on data from MIDUS I (Midlife in the United States: A National Study of Health and Well-being) and its 10-year mortality follow-up. Results show that health differences between 10-year survivors and non-survivors explain 56% of their differences in socio-political participation. https://www.sciencedirect.com/science/article/abs/pii/S0277953618300108 https://www.sciencedirect.com/science/article/abs/pii/S02779...

gerdesj 3y ago

The article is peppered with links to studies and quotes. for example: "as a 2017 CDC study showed" - showed is a hyperlink to: https://edition.cnn.com/2017/05/02/health/african-american-death-rates-cdc-study/index.html https://edition.cnn.com/2017/05/02/health/african-american-d... which links to https://www.cdc.gov/vitalsigns/aahealth/index.html https://www.cdc.gov/vitalsigns/aahealth/index.html The article quotes: "For blacks 18 to 64, the data showed that they were at a higher risk of early death than whites." I would not expect a journo piece to feature actual maths but I do want to see sources and we have them. I have no idea what your final sentence really means but it does: "guess". Sources?

nathan_compton 3y ago

The way our first past the post system works, a few percentage points is enormously impactful. A few percentage point swing in Florida could change presidential elections permanently.

storkhm 3y ago

I would think cigarette smoking is a big variable in the effect too. People living in poverty, uninsured, didn't graduate high school are all variables we know correlate with a higher % of cigarette smoking.