Regarding "too broad". I think there are two sorts of ways a question could be too broad. One is the sort of question that amounts to summarizing an entire course or text book: "Please tell me all about regression and how I can do it on my data. Go step by step." That seems clearly way too broad and an easy vote to close.
But there are other questions that ask for "general recommendations" that, I think, can be answered briefly in ways that are often helpful to the questioner. "What are some ways I might analyze these data?" and the answer might be "try XXX which does XXXX, also try YYY and ZZZ" or even a list of methods, without a whole lot of detail.
What do others think?