Monday, 28 April 2014

LCI14 Conference proceedings Emil Kirkegaard


Immigration in Denmark and Norway

Emil O. W. Kirkegaard

I acquired data from the official Danish statistics agency for 71 immigrant groups by country of origin. Then I looked at crime rates, educational attainment, income and use of social benefits among the groups. There were large differences. The results show that performance on one socioeconomic measure correlated highly with performance on the other socioeconomic measures and that there was a strong general socioeconomic factor.

(Below is the link to the whole presentation)


I used regression analyses to see how well the group differences were predictable from national IQs, GDP and Islam prevalence in the home country. Surprisingly, Islam was a best predictor for all areas, while national IQs and GDP were also good but not independent predictors. Islam and IQ/GDP are combinable in multiple regression to yield very good predictions at the group level, with a R2 adjusted around 0.7 -- very high for sociology.

Lastly, results showed that the groups that did well tended to do so for both countries, as performance in one country was a good predictor of performance in the other country.

The presentation is based on three papers two of which have already been published in the open access, mandatory data sharing journal Open Differential Psychology. Some of the data is from a paper in review, also in the same journal. As the datasets are public, curious readers might want to try their own preferred predictors of socioeconomic measures for Denmark and Norway.


Emil Ole William Kirkegaard, stud.cand.mag (Aarhus University, linguistics) personlig hjemmeside / personal homepage.

1 comment:

  1. Very interesting Emil. You hypothesise that some of the correlation between the proportion of Islamic migrants and crime rates is attributable to relatively low IQ of those migrants, but couldn't the causation also run the other way: the citizens of societies with low average IQ are naturally attracted to unsocial activities, including the more violent aspects of Islam. Hence there is a statistical identification problem.