The National Indigenous Reform Agreement aims to:
• increase the proportion of Indigenous children who are enrolled in and attending a preschool program in the year before formal schooling
• halve the gap in reading, writing and numeracy for Indigenous students within a decade (by 2018)
• at least halve the gap for Indigenous students in year 12 attainment or equivalent attainment rates by 2020
• halve the gap in employment outcomes between Indigenous and non Indigenous Australians within a decade (by 2018)
The picture on the cover tells you why this program is bound for success — eventually.
Following up with the last post I decided to look at the correlation between exogamy and the magnitude of the gap. As an index of cognitive ability I used NAPLAN math. Since I have exogamy rates by region and geolocation for 2001, I looked at grade 5 (average age 10.6) 2011 scores. Row A lists the region; B the # of Aborigines; C the standardized difference; D the 2001 exogamy rates; E the inferred % Australoid for the parental generation; F the inferred % Australoid for the offsprings; G/H the predicted d if there originally was a geneotypic d of 2; I the difference between prediction and observation. (To note, I used the 2001 exogamy rates to infer parental admixture and double checked this with the estimates based on genotyping. For example, McEvoy et al (2010) reports that Aborigines (presumably, adults) in NSW are 36% European — and this is approximately what we get if assume that the current rate of introduction of European genes into the Aboriginal gene pool, via contemporaneous admixture, (68% x 0.5 in NSW) approximates the total previous admixture. There are a lot of assumptions here, of course, but errors should attenuate the correlation between regional differences in admixture and gaps, which was 0.96.
I repeated the same analysis using geolocation (Metro, provincial, remote areas, very remote areas). Here, the correlation was much lower, but not trivial, at about .65. (I only had exogamy data for the regional capitals and for all other regional locations combined. I assumed that the former was a fair proxy for the rates in Metro areas (and maybe provincial areas too) and that the latter was an OK proxy for rates in remote and very remote areas).
Here: A gives the region,; B the score difference in Metro areas; C the inferred % Australoid in Metro regions (of the students); D the correlation between differences in scores and admixture in Metro areas (B,C); E through G the score differences in Provincial, Remote, and Very Remote areas; H the mean of E through G; I the inferred admixture in these regions combined; J the correlation between differences in scores and admixture in these combined regions (H,I); K the difference in gaps between Metro and other locations; I the difference in admixture between Metro and other locations; M the correlation between K and L; N1 same as K but lumped Provincial with Metro; N2 correlation between L and N; O and Q, the predicted differences assuming a 2 SD genotypic difference; P and R difference between predicted and found difference (B-O) and (H-Q).
So, for whatever reason, there seems to be a spatial association between Australoidness and the magnitude of differences. One could imagine numerous pathways by which this would occur, of course. Obviously one would want to check if there was a cross cohort difference (e.g., what the magnitude of the gap was, per region, in the 1970s). Whatever the case, over the coming years, Chuck predicts particularly stubborn gaps in NT and WA especially deep in the boondocks where the Autraloid blood runs more pure.
NALPLAN online explorer
Population Distribution, Aboriginal and Torres Strait Islander Australians, 2006
Head et al., 2010. INTERMARRIAGE IN AUSTRALIA:Patterns by birthplace, ancestry, religion and indigenous status
McEvoy et al., 2010. Whole-Genome Genetic Diversity in a Sample of Australians with Deep Aboriginal Ancestry