Erik Moeller wrote:
"Pennsylvania German" is just not a term in use.
Try googling for it.
"Pennsylvania German" returns 17,200. "Pennsylvania Dutch" returns 64,800.
What matters is not just the number of sites, but also the type of sites.
That's right, I think.
I don't personally think that there is any *simple* rule we can use like "Always trust Google counts" or "Always trust Britannica" or "Always trust academics" or "Always trust the dictionary" or "Never use fancy letters with diacriticals". Each of those things counts as a _factor_ in determining what we should prefer, and none of them will always trump the others.
If all that Erik is saying is that there are cases where "academically correct" should trump "popular usage" then I would agree. But there are lots of cases where "popular usage" should trump "academically correct".
In this case, it is possible that [[Pennsylvania Dutch]] should be an article that starts off by pointing out that it's a popular but inaccurate term, pointing people to [[Pennsylvania German]] for an actual history of the people, culture, etc. And then the rest of the page could be a history of the _term_.
But I don't think any simple rule is possible.
--Jimbo