Looking into the paper (see the sci-hub link), it's a bit odd. Data source is Mechanical Turk respondents in the USA, and Danish military draftees in Europe, w

Looking into the paper (see the sci-hub link), it's a bit odd. Data source is Mechanical Turk respondents in the USA, and Danish military draftees in Europe, who had to take a 'cognitive ability test'. "Political groups" don't include members of political parties - they have 'neo-Nazis', 'Christian fundamentalists', 'socialists' and 'communists'. Seems like there are a great many other groups one could list. Social science papers are notorious for being non-replicabale and based on non-representative samples, so I doubt these conclusions really mean much.