So, I posted this:
Abstract
We used data from the PING study (n≈1200) to examine the relationship between cognitive ability, socioeconomic outcomes and genomic racial ancestry. We found that when genomic ancestry was not included in models, self-reported race/ethnicity (SIRE) was a useful predictor of cognitive ability/S, but when genomic ancestry was included, SIRE lost much or most of its validity. In particular, for African Americans, the SIRE standardized beta changed from about -1.00 to .20. European genomic ancestry was found to be positively related to cognitive ability/S (r’s .26/.33) including when SIRE was controlled, while African genomic ancestry was found to be negatively related to cognitive ability/S (r’s-.36/-.30) also when SIRE was controlled.Key words: race differences, group differences, intelligence, IQ, cognitive ability, social status, inequality, ancestry, admixture, genomics
It’s note quite definitive because it lacks ability to distinguish between colorism and genetic effects. Environmentalists are very stubborn, so one needs very strong evidence. So, we need datasets that have:
- Genomic ancestry, preferably not using just a few AIMs
- Racial appearance data: skin brightness in particular
- Cognitive ability
- Socioeconomic indicators (for S factor)
What datasets are there like this? Let’s find them and start applying so we can settle this question once and for all.
Pelotas (Brazil) Birth Cohort Study
- n=3700.
- Mixed race.
- All four variables of interest.
- Requires approval.
- In Portuguese.
The Coronary Artery Risk Development in Young Adults Study (CARDIA)
- “It began in 1985-6 with a group of 5115 black and white men and women aged 18-30 years”
- 3 cognitive tests: Rey Auditory-Verbal Learning Test (RAVLT), Digit Symbol Substitution Test (DSST) and Stroop Test
- Skin brightness. Can’t see on the website, but used in studies that used the dataset.
- Socioeconomic outcomes. Yes, same as above.
- Requires approval.
UK Biobank
- n=500k planned. Currently about 100k.
- Mostly Britons, but some British Africans and others.
- Crappy cognitive tests, good genomic ancestry and no racial phenotype data (I think).
- Requires approval. Must be for medical reasons.
The Multi-Ethnic Study of Atherosclerosis (MESA)
- “a diverse, population-based sample of 6,814 asymptomatic men and women aged 45-84 from six field centers across the United States. Approximately 38 percent of the recruited participants are white, 28 percent African-American, 22 percent Hispanic, and 12 percent Asian, predominantly of Chinese descent.”
- Some, but unknown cognitive ability data.
- Socioeconomic indicators.
- Requires approval.
- Self-perceived discrimination.
- Undeclared anthropometric measures, possibly skin brightness.
- Genomic ancestry? There is a person in charge of ‘genetic ancestry’.
Add Health
- n≈15,000
- n≈2,000 with genomic data
- Cognitive data: PPVT
- Educational attainment, parental S data
- Skin color [1-5]
