> These results are not comprehensive, however. Nor are they representative. EF’s index is based on the results of a free online test taken by 2.3m volunteers in 100 countries. Only people with an internet connection and time and willingness to take a test are included in the sample...
I almost stopped reading at this point.
> Such biases aside, the EF’s index produces results that are interesting...
I was taught that in scientific research one should never claim that her/his own research is interesting, not least because it's something that's up to the audience to decide.
> ... interesting, if not entirely scientific.
At least they are honest about it not being entirely scientific.
> This correlates with another factor: EF repeatedly finds that English skills are highly correlated with connections and openness to the rest of the world.
Openness...? By what measure?
Edit: typo.