The data is quite old, so the number of speakers has probably increased drastically since then. Still, I guess that the proportions and geographical distribution should be more or less the same today.
The data is quite old, so the number of speakers has probably increased drastically since then. Still, I guess that the proportions and geographical distribution should be more or less the same today.
Other than that this particular one is a bit ugly I don’t see an issue here. I agree in other cases its often used to diffuse differences between different sets if data but here its kept relatively simple in my opinion.