[pdeuchler]

Comparing statistics from small, homogeneous populations to statistics from larger, heterogeneous populations is fraught with peril. Doubly so for metrics on education and health. (I like to call this the Scandinavian Fallacy)

As engineers we would never compare uptime statistics from a small, niche startup serving a couple thousand people to uptime statistics from, say, Google. Why do we immediately ignore these principles outside of engineering?

Edit: wording

[tomp]

I'd expect the uptime of Google to be better than the uptime of a niche startup.

[Hello71]

small, niche startups usually have poor uptime on their services. we commend them if they have a year up (hell, if they still exist after a year), but criticize Google for being down once every few years.

therefore, if you want to bring in the analogy of startups and Google, it should be even more impressive that small countries (viz. people in small countries, viz. Cuba) live longer than Google (viz. people in large countries, viz. United States).

[EpicEng]

No, because the specific topics at hand are not comparable (service uptime and, say, physicians per capita). The analogy was meant only to illustrate the fact that you cannot (should not) directly compare statistics from such different entities without attempting to control for the variables.

[avar]

What does the heterogeneous populations of the US have to do with it? Maybe when it comes to life expectancy you're right, because genetics is a large factor in that. But when it comes to literacy rate and the proportion of physicians that's entirely down to government policy.

Do you really think the US couldn't beat Cuba on those metrics if they decided spending money on those topics was more important than spending money on the military?

[aaron-lebo]

I believe the point that is being made is that the US was designed from the outset with the idea that diverse factions compete against factions, and that was the only way such a large and varied democracy could work. See Federalist 10.

Citizens of the US have wildly varied views about what it means to be a good American, and we often have greater cohesion to groups such as race or state, etc. Just about the only thing we do agree upon is when there is an existential threat to the country, hence military growth over time.

Cuba or Norway have much less divided societies to govern, and Cuba's leadership is so small that it can focus on specific issues in a way US politics cannot.

[avar]

Sure, that's a factor in US politics. I'm just saying that has nothing to do with having a heterogeneous population as the parent was claiming.

The parent was comparing the proportion of physicians in Cuba v.s. the US and claiming that the US couldn't match Cuba because of its heterogeneous population.

Now if you look at the WHO report on density of physicians[1] you see first-wold countries like Japan and Canada ranking below the US on the number of physicians per-capita.

I think numbers like these have a lot more to do with how the health care system is structured than the sort of population you have. E.g. maybe nurses in Japan and Canad have a bigger role in health care than the US.

1. http://www.who.int/gho/health_workforce/physicians_density/e...

[baddox]

Government policy is heavily affected by cultural heterogeneity.

[finid]

> Comparing statistics from small, homogeneous populations

Cuba is small. Does it have an homogeneous population?

No, amigo!

[sdm]

Since when has Cuba had a homogeneous population?

[aaron-lebo]

I really like that term. Is that your own?