{"id":14032,"date":"2025-04-08T16:20:13","date_gmt":"2025-04-08T15:20:13","guid":{"rendered":"https:\/\/emilkirkegaard.dk\/en\/?p=14032"},"modified":"2025-04-08T16:29:07","modified_gmt":"2025-04-08T15:29:07","slug":"worldwide-subnational-intelligence-estimates","status":"publish","type":"post","link":"https:\/\/emilkirkegaard.dk\/en\/2025\/04\/worldwide-subnational-intelligence-estimates\/","title":{"rendered":"Worldwide subnational intelligence estimates"},"content":{"rendered":"

It is already known that intelligence causes social status for persons. There’s a very large amount of cross-sectional data, and causal conclusions are derived from various family studies, such as sibling studies or parent-child. These show that the smarter sibling tends to do better than the duller siblings<\/a>, and the same is true for parent-child comparisons. These have been augmented by polygenic scores in recent years with the same findings<\/a>:<\/span><\/p>\n

We tested if education-linked polygenic scores predicted social mobility in >20,000 individuals in five longitudinal studies in the United States, Britain, and New Zealand. Participants with higher polygenic scores achieved more education and career success and accumulated more wealth. However, they also tended to come from better-off families. In the key test, participants with higher polygenic scores tended to be upwardly mobile compared with their parents. Moreover, in sibling-difference analysis, the sibling with the higher polygenic score was more upwardly mobile. Thus, education GWAS discoveries are not mere correlates of privilege; they influence social mobility within a life.<\/p><\/blockquote>\n

It is also known that at the national level, intelligence also correlates with better outcomes. Here causality is harder to test, mainly because we cannot rely on siblings and cannot experimentally change intelligence, but still we can do kitchen sink regression models, letting intelligence compete with every alternative variable people have come up with. There are two such studies, both of which found that intelligence is the key predictor<\/a>, along with being poor to begin with (we might add, not having communism<\/a>).<\/span><\/p>\n

Between individuals and nations, there are a number of possible intermediate aggregations we could look at. The most common choice is subnational units of some country. These come in various levels, as each country uses its own hierarchy or quasi-hierarchy. These are called administrative divisions<\/a> and there may be 5+ levels for a given country.<\/p>\n

\"\"<\/p>\n

So for USA, the first level is states (50), the second level is counties (3000+). Things can get complicated because of special units that may break the clean structure in certain ways. The US has a federal district, which mostly acts like a state (DC), but it also has 300+ Indian reservations, unincorporated territories (e.g. Guam) and other units (e.g. <\/span>Minor Outlying Islands<\/span><\/a>). Some countries seem to take pride in making things as complicated as possible. Australia has 6 states, 3 internal territories, and 7 external territories. Each of these have a bewildering variety of smaller units like shires, cities, councils and so on. In general though, each state orders them hierarchically, so that the nation unit is split into a number of first-order administrative divisions, then these split into second-order divisions and so on, and even if units have different names, they are still in the same level of the hierarchy.<\/span><\/p>\n

Some effort has been made in the last 20 years to conduct studies of subnational units with intelligence and socioeconomic measures. Our 2018 meta-analysis tried to include all of these:<\/span><\/p>\n