scottlowther wrote:Our life expectancy is the best in the world (or at least among the very best) when you factor out violent deaths, which no health care system can help with. As pointed out int he references, a good chunk of our infant mortality is due not to the health care system but to the behaviors of the (often way-too-young) mothers. Our nation is far more ethnically diverse than the others at the top of the list... and like it or not, that seems to matter for reasons divorced from the health care system.
Violence in a society is also indicative of it's social ills. Therefore, I can hardly see a reason to exclude violent deaths from the statistics of life expectancy. If anything, they reflect the state a society is in all the better.
A superviolent society will not have a high life expectancy, which is natural, and which is damning if you think the only reason for the US ranking so bad is it's violence. That means the US is a hyperviolent society compared to other First World nations.
But hey, I'm a "left-wing extremist", so what would I know about statistics (despite having a 2-year course of general theory of statistics and socio-economic statistics during my formal education), yup?
Lì ci sono chiese, macerie, moschee e questure, lì frontiere, prezzi inaccessibile e freddure
Lì paludi, minacce, cecchini coi fucili, documenti, file notturne e clandestini
Qui incontri, lotte, passi sincronizzati, colori, capannelli non autorizzati,
Uccelli migratori, reti, informazioni, piazze di Tutti i like pazze di passioni...
...La tranquillità è importante ma la libertà è tutto!
scottlowther wrote:1) Governemnt governs best when it governs least
The full Walter Lippmann quote, the second half of which somehow always gets snipped off by conservatives and libertarians, reads: "It is perfectly true that the government is best which governs least. It is equally true that the government is best which provides most."
When ballots have fairly and constitutionally decided, there can be no successful appeal back to bullets.
—Abraham Lincoln
People pray so that God won't crush them like bugs.
—Dr. Gregory House
Oil an emergency?! It's about time, Brigadier, that the leaders of this planet of yours realised that to remain dependent upon a mineral slime simply doesn't make sense.
—The Doctor "Terror Of The Zygons" (1975)
Stas Bush wrote:
Violence in a society is also indicative of it's social ills. Therefore, I can hardly see a reason to exclude violent deaths from the statistics of life expectancy. If anything, they reflect the state a society is in all the better.
Also, it's not exactly honest to say a health care system can't do anything violent deaths: a good health care system will, in fact, shove some victims from the "dead" tally into the "victimized by violence" tally. Then there's other government activities that contribute to reduction of violent deaths, like quality of emergency services, overall crime rate which either prevent or increade the incidents of violence themselves.
Excluding violent deaths from a statistical measure and then using such adjusted statistics to compare two countries is like comparing profits of corporations A and B while throwing out costs of debt servicing for A, and then declaring B is better managed because of lower costs.
Incidentally, this is also why HDI is a far superior measure of the quality of life than pure life expectancy.
JULY 20TH 1969 - The day the entire world was looking up
It suddenly struck me that that tiny pea, pretty and blue, was the Earth. I put up my thumb and shut one eye, and my thumb blotted out the planet Earth. I didn't feel like a giant. I felt very, very small. - NEIL ARMSTRONG, MISSION COMMANDER, APOLLO 11
Signature dedicated to the greatest achievement of mankind.
MILDLY DERANGED PHYSICIST does not mind BREAKING the SOUND BARRIER, because it is INSURED. - Simon_Jester considering the problems of hypersonic flight for Team L.A.M.E.
What is the dependency of the Human Development Index on life expectancy?
And, come to think of it, does it also have a separate dependency on infant mortality? Infant mortality rates have a huge effect on life expectancy, because averaging in even a small number of people who died at the age of one will offset a large number of people who live to sixty, seventy, or eighty.
Simon_Jester wrote:What is the dependency of the Human Development Index on life expectancy?
And, come to think of it, does it also have a separate dependency on infant mortality? Infant mortality rates have a huge effect on life expectancy, because averaging in even a small number of people who died at the age of one will offset a large number of people who live to sixty, seventy, or eighty.
A third of the weighting comes from life expectancy- infant mortality is included in that and not counted seperately. So yes, a high infant mortality will drag it down.
Samuel wrote:A third of the weighting comes from life expectancy- infant mortality is included in that and not counted seperately. So yes, a high infant mortality will drag it down.
OK, but my question was a little different. I was asking if the HDI depended on infant mortality above and beyond its dependency on life expectancy.
Imagine two countries with identical overall life expectancy, and that are identical in all other ways, but that have different infant mortality rates (adults live longer in the country with higher infant mortality to balance out life expectancy). Would the two countries have different HDIs? It appears the answer is "no," which means that I'm satisfied.
If the answer were "yes," then infant mortality would have an even stronger effect on the HDI than it does on life expectancy, which might be giving it even more than the (high) weight it deserves.
The HDI statistic was created because it is composed of easily available data, even in the poorest countries. To be able to compare virtually any country in the world. However, it was never meant to be the be all end all statistic to determine what country is the "best".
Simon_Jester wrote:What is the dependency of the Human Development Index on life expectancy?
And, come to think of it, does it also have a separate dependency on infant mortality? Infant mortality rates have a huge effect on life expectancy, because averaging in even a small number of people who died at the age of one will offset a large number of people who live to sixty, seventy, or eighty.
US life expectancy is small relative to other developed countries because Americans eat more, eat badly and hence, become fatter and life less than other populations with similar level of economic development.
Japan ----------- 78.7
United Kingdom -- 75.6
Germany -------- 75.4
US -------------- 75.3
France has a high per capita calorie intake and a high life expectancy. That's because they consume healthier types of food, than the Anglo Germanic peoples.
Simon_Jester wrote:adults live longer in the country with higher infant mortality to balance out life expectancy
It's very hard to fully balance it out - the longer lives of adults offer incremental gains, while child mortality drags down LE at birth like an anchor. Not that it's impossible, but I probably don't know of any examples.
Lì ci sono chiese, macerie, moschee e questure, lì frontiere, prezzi inaccessibile e freddure
Lì paludi, minacce, cecchini coi fucili, documenti, file notturne e clandestini
Qui incontri, lotte, passi sincronizzati, colori, capannelli non autorizzati,
Uccelli migratori, reti, informazioni, piazze di Tutti i like pazze di passioni...
...La tranquillità è importante ma la libertà è tutto!
Alyrium Denryle wrote:Not necessarily. Only when you are taking a sample. There are only error bars if you are sampling say, 2000 out of the entire population and creating a 95% confidence interval for the mean. If life expectancy is calculated for the whole population and is thus the parametric mean for life expectancy. There is no error. Of course mean life expectancy is a projection so there should be error bars.
Infant mortality rate however is definitely the parametric.
Okay, that's a fair point. Still, life expectancy (in particular), even when calculated for the entire population, is going to follow a distribution, and it's important to know the spread of the distribution as well as the mean in order to know whether there is a significant (but not statistically significant) difference.
A Government founded upon justice, and recognizing the equal rights of all men; claiming higher authority for existence, or sanction for its laws, that nature, reason, and the regularly ascertained will of the people; steadily refusing to put its sword and purse in the service of any religious creed or family is a standing offense to most of the Governments of the world, and to some narrow and bigoted people among ourselves.
Simon_Jester wrote:adults live longer in the country with higher infant mortality to balance out life expectancy
It's very hard to fully balance it out - the longer lives of adults offer incremental gains, while child mortality drags down LE at birth like an anchor. Not that it's impossible, but I probably don't know of any examples.
Oh, I know; in fact I pointed that out myself a few posts ago. I was using it as a thought experiment to illustrate my question about how the HDI depends on infant mortality.
I was also thinking in terms of infant mortality rates in developed or near-developed countries. For instance, the difference between 0.5% and 1% infant mortality is only going to knock several months off life expectancy, which theoretically could be balanced out by lifestyle differences... somehow. Not likely to happen, of course, for reasons that are fairly obvious...
Iosef Cross wrote:US life expectancy is small relative to other developed countries because Americans eat more, eat badly and hence, become fatter and life less than other populations with similar level of economic development. [tables follow]
Umm... okay.
I didn't actually ask this question, and am not surprised by the answer (though I am surprised by the sheer magnitude of our per capita calorie intake). Originally, I was just asking a very hypothetical question about the relationship between the HDI and infant mortality rates.
My point is that the discrepancy of life expectancy between the US and Europe & Japan can be explained by diet, instead of statistical manipulations of infant mortality and bad health systems.
The fact is that the main determinants of life expectancy are vaccination, access to good water sources, good living conditions (i.e.: absence of exposure to toxic chemicals, decent number working hours) and diet. Japanese and French live longer than anybody else because they have healthy diets compared to the Anglo Germanic countries, with have comparable access to vaccination, water sources and good living conditions.
Americans without much money and health insurance cannot afford costly cancer treatment or heart transplants, but these types of treatment have minimal impact in life expectancy.
Iosef Cross wrote:My point is that the discrepancy of life expectancy between the US and Europe & Japan can be explained by diet, instead of statistical manipulations of infant mortality and bad health systems.
All right, though since I did not attempt to perform such a manipulation in order to explain the discrepancy, it seemed odd as a response to my post.
I don't disagree with your argument; it just seemed kind of random to me.
Alyrium Denryle wrote:Not necessarily. Only when you are taking a sample. There are only error bars if you are sampling say, 2000 out of the entire population and creating a 95% confidence interval for the mean. If life expectancy is calculated for the whole population and is thus the parametric mean for life expectancy. There is no error. Of course mean life expectancy is a projection so there should be error bars.
Infant mortality rate however is definitely the parametric.
Okay, that's a fair point. Still, life expectancy (in particular), even when calculated for the entire population, is going to follow a distribution, and it's important to know the spread of the distribution as well as the mean in order to know whether there is a significant (but not statistically significant) difference.
Well, if you have the entire population, the differences will be significant. Your statistical power when you have the whole population can best be summed up by the word "Yes"
When you run a T Test on two samples of a population to determine an effect of treatment you are comparing their means and distributions in order to determine if they come from the same underlying distribution and the differences you see are due to chance, or if there is a treatment effect and they could not have come from the same distribution with a high probability. If you are comparing two populations and there is a difference, then it is definitely significant. Now, that brings us to the big question. Are two countries the same population (all of humanity) or are they two? You could make a good case for the former, but even then the sample sizes are so huge that statistical power is approaching 100%. A statistically significant difference is almost a certainty even with very small effect sizes.
GALE Force Biological Agent/
BOTM/Great Dolphin Conspiracy/ Entomology and Evolutionary Biology Subdirector:SD.net Dept. of Biological Sciences
There is Grandeur in the View of Life; it fills me with a Deep Wonder, and Intense Cynicism.
Isn't American diet crappy because of the subsidization of sugar and corn syrup? I know that those foods taste good, but if they were so cheap- or if unhealthy stuff was taxed- people would probably be fitter.
Of course, the hilarity is that even if American health care is as effective as all the other rich countries we still pay massively more for it.
Iosef Cross wrote:My point is that the discrepancy of life expectancy between the US and Europe & Japan can be explained by diet, instead of statistical manipulations of infant mortality and bad health systems.
The fact is that the main determinants of life expectancy are vaccination, access to good water sources, good living conditions (i.e.: absence of exposure to toxic chemicals, decent number working hours) and diet. Japanese and French live longer than anybody else because they have healthy diets compared to the Anglo Germanic countries, with have comparable access to vaccination, water sources and good living conditions.
Americans without much money and health insurance cannot afford costly cancer treatment or heart transplants, but these types of treatment have minimal impact in life expectancy.
There is a way to fix this dilemma. Variance partitioning via a multiple regression. Unfortunately I dont have a dataset handy for obesity statistics, but I may be able to find per capita food consumption in calories to use as a proxy.
GALE Force Biological Agent/
BOTM/Great Dolphin Conspiracy/ Entomology and Evolutionary Biology Subdirector:SD.net Dept. of Biological Sciences
There is Grandeur in the View of Life; it fills me with a Deep Wonder, and Intense Cynicism.
Japan and USA sure have extremely different calorie consumption - the gap between USA and Japan rose from ~400 in 1961 to ~1000 in 2002 - while the average Japanese consumption still remains in the 2500+ range, the U.S. is well over 3000.
Americans consume more calories per capita, but that is not, in and of itself, a pattern of unhealthiness - European nations like France, Germany and Finland (which I picked off-hands) consume in the 3000+ range. Germany and France have nigh identical PC calorie consumption.
The healthiness of the calories themselves must be called into question, when simple levels do not explain enough. A large fraction of the calories consumed in the US is junk foods. I'm not sure if anyone did solid calculations on how much junk foods do people in the USA consume compared to other industrialized nations (that'd be a quite worthwhile study!).
Lì ci sono chiese, macerie, moschee e questure, lì frontiere, prezzi inaccessibile e freddure
Lì paludi, minacce, cecchini coi fucili, documenti, file notturne e clandestini
Qui incontri, lotte, passi sincronizzati, colori, capannelli non autorizzati,
Uccelli migratori, reti, informazioni, piazze di Tutti i like pazze di passioni...
...La tranquillità è importante ma la libertà è tutto!
Iosef Cross wrote:Yes, Germans live considerably less than French even though they eat similar quantities of calories because the French has a healthier diet.
You know, what really baffles me about his statements is the blindness depicted.
If you take your own country as baseline, and find EVERY other country to be more 'socialleftist' than your own, the REASONABLE thing to deduct is that your country is an right wing extremist, not that it's the only centrist still left on the field.
A minute's thought suggests that the very idea of this is stupid. A more detailed examination raises the possibility that it might be an answer to the question "how could the Germans win the war after the US gets involved?" - Captain Seafort, in a thread proposing a 1942 'D-Day' in Quiberon Bay
LaCroix wrote:You know, what really baffles me about his statements is the blindness depicted.
If you take your own country as baseline, and find EVERY other country to be more 'socialleftist' than your own, the REASONABLE thing to deduct is that your country is an right wing extremist, not that it's the only centrist still left on the field.
"But we can't be fascist/right wing. After all, we defeated them in the glorious war. We would never become like people we defeated several generations ago."
Whoever says "education does not matter" can try ignorance
------------
A decision must be made in the life of every nation at the very moment when the grasp of the enemy is at its throat. Then, it seems that the only way to survive is to use the means of the enemy, to rest survival upon what is expedient, to look the other way. Well, the answer to that is 'survival as what'? A country isn't a rock. It's not an extension of one's self. It's what it stands for. It's what it stands for when standing for something is the most difficult! - Chief Judge Haywood
------------ My LPs
Iosef Cross wrote:Americans without much money and health insurance cannot afford costly cancer treatment or heart transplants, but these types of treatment have minimal impact in life expectancy.
They cannot afford many types of treatment for cardiovascular diseases, including routine observation which is critical for preventive care.
Until I see a solid statistical demonstration that there is no correlation between heart disease death rates, accessibility of care and average life expectancy. There is solid correlation between heart disease deaths and ALE in Russia.
So one should look if accessibility of care impacts cardiovascular diseases, which are by and large a very huge factor in modern mortality.
Lì ci sono chiese, macerie, moschee e questure, lì frontiere, prezzi inaccessibile e freddure
Lì paludi, minacce, cecchini coi fucili, documenti, file notturne e clandestini
Qui incontri, lotte, passi sincronizzati, colori, capannelli non autorizzati,
Uccelli migratori, reti, informazioni, piazze di Tutti i like pazze di passioni...
...La tranquillità è importante ma la libertà è tutto!
Surlethe wrote:Okay, that's a fair point. Still, life expectancy (in particular), even when calculated for the entire population, is going to follow a distribution, and it's important to know the spread of the distribution as well as the mean in order to know whether there is a significant (but not statistically significant) difference.
Well, if you have the entire population, the differences will be significant. Your statistical power when you have the whole population can best be summed up by the word "Yes"
When you run a T Test on two samples of a population to determine an effect of treatment you are comparing their means and distributions in order to determine if they come from the same underlying distribution and the differences you see are due to chance, or if there is a treatment effect and they could not have come from the same distribution with a high probability. If you are comparing two populations and there is a difference, then it is definitely significant. Now, that brings us to the big question. Are two countries the same population (all of humanity) or are they two? You could make a good case for the former, but even then the sample sizes are so huge that statistical power is approaching 100%. A statistically significant difference is almost a certainty even with very small effect sizes.
I think we're talking crosswise. I mean that the population distributions might be different, but not "significantly" (not in the statistical sense) different. Say the US has a mean life expectancy of 78 and France has a mean life expectancy of 80. If the US standard deviation of life expectancy is 6 years, then there's not much of a difference between US life expectancy and Japanese life expectancy.
* The actual numbers don't matter for the sake of argument.
A Government founded upon justice, and recognizing the equal rights of all men; claiming higher authority for existence, or sanction for its laws, that nature, reason, and the regularly ascertained will of the people; steadily refusing to put its sword and purse in the service of any religious creed or family is a standing offense to most of the Governments of the world, and to some narrow and bigoted people among ourselves.
Surlethe wrote:Okay, that's a fair point. Still, life expectancy (in particular), even when calculated for the entire population, is going to follow a distribution, and it's important to know the spread of the distribution as well as the mean in order to know whether there is a significant (but not statistically significant) difference.
Well, if you have the entire population, the differences will be significant. Your statistical power when you have the whole population can best be summed up by the word "Yes"
When you run a T Test on two samples of a population to determine an effect of treatment you are comparing their means and distributions in order to determine if they come from the same underlying distribution and the differences you see are due to chance, or if there is a treatment effect and they could not have come from the same distribution with a high probability. If you are comparing two populations and there is a difference, then it is definitely significant. Now, that brings us to the big question. Are two countries the same population (all of humanity) or are they two? You could make a good case for the former, but even then the sample sizes are so huge that statistical power is approaching 100%. A statistically significant difference is almost a certainty even with very small effect sizes.
I think we're talking crosswise. I mean that the population distributions might be different, but not "significantly" (not in the statistical sense) different. Say the US has a mean life expectancy of 78 and France has a mean life expectancy of 80. If the US standard deviation of life expectancy is 6 years, then there's not much of a difference between US life expectancy and Japanese life expectancy.
* The actual numbers don't matter for the sake of argument.
I just calculated it using a paired T Test for Means. 76.6 for the US, 81.1 for japan, assuming SD of 6. They are highly significant. P value of <<<<<<.00001
As I said, the power to detect very tiny effect sizes even with large standard deviations with a sample size in the millions is basically Yes.
GALE Force Biological Agent/
BOTM/Great Dolphin Conspiracy/ Entomology and Evolutionary Biology Subdirector:SD.net Dept. of Biological Sciences
There is Grandeur in the View of Life; it fills me with a Deep Wonder, and Intense Cynicism.
So what are the US and Japanese population stdevs?
A Government founded upon justice, and recognizing the equal rights of all men; claiming higher authority for existence, or sanction for its laws, that nature, reason, and the regularly ascertained will of the people; steadily refusing to put its sword and purse in the service of any religious creed or family is a standing offense to most of the Governments of the world, and to some narrow and bigoted people among ourselves.
