This strikes me as an abuse of statistics. The author seriously lacks data (only 36 data points!) and seems perfectly content with sneaking in causal claims whe

This strikes me as an abuse of statistics. The author seriously lacks data (only 36 data points!) and seems perfectly content with sneaking in causal claims when the regressions plainly don't support them. Most of these correlations seem to be driven largely by 3 outliers-- Turkey and Mexico on the negative end, and Norway on the positive end. Just look at this graph, remove those two or 3 points, and imagine performing the regression again. Big difference. [0] What baffles me is for the relationship between socialism and GDP/hour worked, the author plainly admits the r-squared is weak but passes it off as powerful evidence anyway: "While there is a lot of variance around the trendline (r^2=0.29), the trend goes in the opposite direction of what the skeptics would argue" I think this is a valuable area of research to explore, but the author is peddling weak correlations to bolster a dubious causal claim. [0]https://lh4.googleusercontent.com/SUvlyxDYd4DzfC4auF6vhBFSXjx6l5fiTsEXkMRtN8MPkG2gNFU_-FCgBMNXcU7i9Xzud4S1AVCerzHHTT56MUF1XvQLVJ-LO3WSyYZ3_4LY-DhPA3GMX6vgFvN-pEtjVLBwRMYM https://lh4.googleusercontent.com/SUvlyxDYd4DzfC4auF6vhBFSXj...