Strategies for answering questions that may be unclear

Question

I know that the common and most acceptable way to deal with poorly posed questions is to point the OP to ways to better ask questions and to probe for clarification in comments. This issue comes up a lot and it is not always the OPs fault. We want people to participate when they are interested in statistics even if their background is limited. Also this is a global community and many community members that ask questions have to do it in English when English is not their first language.

Sometimes I ask a lot of probing questions such as how I helped with Joe King's question today. But other times the question is just missing something. it is ambiguous but I can make an educated guess at the OPs intended question. I choose to answer the what I think the intended question is. If i am right we may have the correct solution much faster than to wait for some clarification(s) to be made. If I am wrong the OP gets to see how I have misinterpreted the question and sometimes can give give the appropriate clarification through amending the question. When this second possibility happens I can delete my answer or amend it to answer the right question. i think this is a good approach. Their could be other less favorable scenarios that can happen but my experience is that this seems to work well.

Nevertheless, I have been criticized by more than one person for answering too quickly.
My question is: Do you agree with my position or do you see something wrong to my approach. If you think it is wrong could you please give your reasoning?

Answer 1

I think one problem with your approach is that people who are not that knowledgeable about statistics may 1) Be asking very unreasonable things, so your educated guesses could be way off, since they assume reasonability and 2) These same people may not recognize when you make assumptions.

I am reminded of a quote from Abraham Lincoln:

If I had 6 hours to chop down a tree, I would spend 4 of them sharpening my axe

With my clients, I often spend a LOT of time figuring out what they want. (Perhaps you do too). I have never found that this time is wasted; indeed, I sometimes still err by not spending enough time doing this.

Answer 2

I don't necessarily have a problem with your default strategy. However, I do want to note (as I have several times on meta), that CV is intended to fulfill multiple goals. One of which is to simply help questioners get answers, and I think your strategy is likely to satisfy that goal (although, @PeterFlom's point that questioners might not be able to recognize that you provided a correct answer to a different question than the one the would have wanted to ask, if they had been able, is worth bearing in mind here). However, another goal is to build a permanent repository of statistical information. Here again, there is some utility to your approach, but it is clearly not as optimal as finding out exactly what the OP actually needs to know, and then providing a comprehensive, clear answer and / or pointing them to other resources where they can get the information they need.