For example, look at these three questions. All tagged reference-request, and all have no answers. That's a major problem. These extremely specific questions can't really be answered unless there's a study that addresses the question, which is pretty unlikely. What should we do about these questions? Downvote them? Close them as too narrow? Embrace them?
Really specific questions aren't bad, they're just hard to answer.
Having very specific questions doesn't cause concern, if every question isn't a super specific reference request that takes a long time to answer.
I'm currently working on an answer to one of those three questions, and I've been working on it for a few hours. That's okay for some questions to take hours to answer, every site has them, but for ALL questions to take hours to answer, that's not good.
Allow the tag to be broader, in terms of what is a credible reference.
The tag wiki currently states:
For questions asking for a published journal to be included as reference in your answer. Please avoid answering questions tagged [reference-request] without referencing a published journal.
But usually, the answer has some other type of source like a website or published study to back up their answer. Like my answer here, though the tag was reference-request, I used two articles as my sources instead. It still worked. Thus the tag seems a little too narrow. Maybe the tag wiki should become:
For questions asking for at least one reliable or published source to be included as a source in the answers.
The community has accepted these questions; the vote counts demonstrate that. I think they can stay.
Do we need a too narrow/too localized/"While similar questions may be on-topic here, this one was resolved in a manner unlikely to help future readers." close reason? Probably.
If the problem is that the request is too specific, the solution would probably be to broaden the question to ask for a request that is less specific, but still relevant to the OP.
"Are there studies on the effect of XYZ method for Chinese speakers learning Russian?" is probably too specific.
"Are there studies on the effect of XYZ method for learning languages?" is probably not too specific, but would still be relevant.
Note that this doesn't mean that a study will actually exist. But a study doesn't need to exist for the question to be properly scoped.