English teachers either are or are not good at teaching.
So it is possible to visualize a contingency table like the following.
Have a doctorate
Not have a doctorate
Good at teaching
Not good at teaching
Which begs the quetion: Is a have-a-doctorate more likely to be good at teaching than a not-have-a-doctorate, other things being equal? The answer is that this is an invalid question to ask. Why? Because other things can never be equal. Fact is that a good teacher, satisfied with their teaching situation, is more likely to NOT opt to venture into something that is so time consuming as a doctoral research. My gut feeling is that those who get their doctorates and those who are good at teaching are drawn from two different populations that overlap very little.