Razib Khan One-stop-shopping for all of my content

October 18, 2010

The “Hispanic Paradox”, and others

Filed under: Culture,data,Data Analysis,Hispanic Paradox — Razib Khan @ 12:04 am

journal.pmed.0030260.g001The New York Times has a piece out on the “Hispanic Paradox”. The paradox is that American Hispanics are longer-lived than non-Hispanic whites, despite the relatively lower socioeconomic status of Hispanics (poorer, less educated). The paradox has been around for a while, and these stories tend to emerge whenever there’s a Census or CDC data release, as there is now. You can get the original data on the CDC website.

But I think one thing about the Hispanic Paradox that needs to be contextualized is that Hispanics are not special in manifesting this paradox. Throwing all non-Hispanic whites into one big pot aggregates real variation. The map on the left is from the paper Eight Americas: Investigating Mortality Disparities across Races, Counties, and Race-Counties in the United States. It shows the life expectancy by county for whites, Hispanic and non-Hispanic. You can see the Hispanic Paradox along the border; most Hispanics in Texas list their race as white, and so show up in these data. But you can also see what I’m talking about in terms of aggregation of non-Hispanic whites for Texas as well. See the counties in the center of Texas with elevated life expectancies? Take a look at this map on the distribution of German ancestry in the USA, and focus on Texas.

More broadly, there are broad swaths of the rural north where whites are relatively poor, but long-lived, in contrast to the South. What you see on the map on the broad purple swaths are the echoes of the Yankee Empire, and the New England Diaspora, which includes the Mormons of Utah. Yankee probity seems to have attracted Scandinavians, Germans, Irish, and Italians, of like mindset. Or the children of immigrants were acculturated to Yankee values.

What is the moral here? Economic development is broadly indicative of life expectancy, but in modern developed societies culture and social milieu matters on the margin, and can swamp the effects of economics. Just as heritability for height is higher in developed societies, so variation in life expectancy within developed societies may be more likely to track cultural categories than economics.

Below are some charts from the original CDC report.




Note: I’m pretty sure that the Hispanic Paradox does not apply to all Hispanics. I am willing to bet that it does not apply to Puerto Ricans. I’ve highlighted the infant mortality data:


Powered by WordPress