Back to [Archive] Other discussions

1
National IQ and % Amerindian/African Ancestry in Latin America
I checked to see if % of African/Amerindian Ancestry predicted indexes of National IQ in Latin America, as Lynn has predicted. The associations were inconsistent and frequently non-existent (especially regarding Amerindian ancestry). Generally, you have countries like Peru and Bolivia (with apparently ~85-90% Amerindian ancestry) who do just as poorly as countries like Paraguay and Uruguay (with apparently 60 to 70% European ancestry, rest Amerindian), etc. Granted, the NIQ indexes -- Lynn's, Altinok's, my GMAT, Age Heaping -- aren't the best; and the admixture studies often aren't representative. (I attached one of the more comprehensive sources for reference.) Any thoughts, though? Should I try breaking e.g., Brazil down by region? It's almost ridiculous to compare Brazil to e.g., Paraguay, given the size differences.
Lynn's IQs for developing countries are not very reliable.
Lynn's IQs for developing countries are not very reliable.


No doubt.
Lynn's IQs for developing countries are not very reliable.


Using Hanushek and Woessmann's Lat. Am. ACH scores (which I just stumbled upon), I find the expected results. After some adjustments to the Admix data, Lynn's total IQ+ACH scores still show a low (negative) r in the case of % Amer, but his IQ-test NIQs yet manage to show a substantial sig. (neg) one. Seems to be a NIQ x ACH effect (which is inflating the total (IQ+ACH) of countries like Bolivia). I'll add color and racial ID to see if I can validate the Admix. estimates.
Admin
Can you post the data? This is a potential good test of simple admixture based theories of national G in admixed South and Latin American countries.

I read recently that the genetics of admixture in these countries are quite odd. Who's to say how this works for admixture for g? Potential for odd effects.

Distant Migration. Although women tend to migrate to the group of their husbands, genetic and historical records suggest that more distant migrations are initiated by men in search of material resources, social status, and reproduc­ tive opportunity (Carvajal-Carmona et al., 2000; Semino et al., 2000). A n example is provided by Carvajal-Carmona et al.’s assessment of m tD N A and Y-chromosome patterns in a Colombian (South America) population that was established by European settlers in the 16th and 17th centuries. The results revealed that the maternal ancestry is largely (more than 90%) Amerindian, whereas the paternal ancestry is largely (94%) European. When combined with the historical record of this population, these genetic patterns paint a picture o f m ale-m ale com petition in which European men displaced Amerindian men to the reproductive benefit of the former and at a large cost to the latter. Related studies have found similar though less extreme patterns in other Amerindian populations (Merriwether et al., 1997) as well as in Melanesia, South Asia, the Middle East, and southern China (Kayser et al., 2003; Quintana-Murci et al., 2004).


The paragraph is from Male, female: The evolution of human sex differences. Second edition. Free ebook here.
I'll have to check the Admix estimates over later; some of them I just eyeballed or guesstimated. The SPSS file contains the following:


Nation
EuAdmix (based Admix studies)
AfAdmix (based Admix studies)
AmAdmix (based Admix studies)

IDWhite (computed based on CIA factbook ethnic ID -- if e.g., reported 50% Mestizo, I recorded this as 25% White and 25% Amerindian)
IDBlack
IDAmer
IDOther

SkinReflect (From G.M)

EuropeanPuttermanWMI (computed based on Putterman's world migration index; I averaged data for European, African, American, and Other nations)
http://www.econ.brown.edu/fac/louis_putterman/world%20migration%20matrix.htm
AmerPuttermanWMI
AfricanPuttermanWMI
OtherPuttermanWMI

HanushekandWoessmann2012
LynnandMalloyIQ
LynnTotalIQ
Altinok2013ACH
EPI
ENGLANGUSAGE
GMAT

SocialProgressIndex2012
HDI2012
I finished with the Ancestry estimates and I threw together a brief discussion:

Racial Ancestry in the Americas. Part 1: Genomic Continental Racial Admixture: Estimate and Validation [Attached, previous attached removed]

For validation, is it OK to use pearson correlation? Is there some other way I should approach this?

The next step will be to see if these ancestry estimates predict anything. But first things first.
1