I have a question about my Cross Validated post: How to set up my workflow and data for my first network model

I would appreciate help. I am trying to ask a question, but I don't understand what I need to do differently. The feedback is make it more focused. Ok... how? I've read the https://stats.stackexchange.com/help/how-to-ask 3 times now.

This is my thinking:

  • Search: well, I've been online all week trying to learn what I need to do, so I think I've gotten this part out of the way.
  • Be on topic: For me, the question asks about setting up an initial dataframe for my network. I think of a set-up as the pseudo-code phase. Plan, make an outline, and try to implement. My post explains my plan and outline. I'm asking if this the correct way to think about it.
  • Be specific: My question is, did I do this right? However, looking back at it, did I ask too many questions pertaining to the one topic I'm talking about? I'm asking about the whole setup, hence the one topic. So should I break each question into a different post?
  • Make it relevant to others: I'm assuming if you aren't interested in the topic, then you don't need to answer.
  • Keep an open mind: That I've definitely got. I am one those people that learn from making mistakes, and I make a lot of them. But I can't fix them unless I understand how to fix them.

So that is what this question is about. Please help me understand what I can do better so that I can ask if my setup to build my network is correct.

You guys have been doing this stuff for a very long time and understand all the aspects to it. I am new to the site, new to programming, new to building a network. It means I don't know what I'm talking about enough to maybe put it into a precise question. I don't know what I have to ask to get the answer I'm looking for. Does that make sense?


2 Answers 2


I didn't vote to close OP's question. But I would have voted to close it in line of needs more focus. So, I am in agreement with the reason it was closed. What did the comment (by whuber) state? It precisely mentioned "[...] to edit this post to narrow it to a specific and definite statistical question."

Why so? Here is my quick take on the present status of the query:

$\bullet$ It is plagued with information not necessarily relevant to a possible question: we really don't need to know whether it emanated from your PhD work and it would expand vastly in the coming years and all that. One can skip the intro, which imo doesn't add anything substantial to the query.

$\bullet$ Then you followed with a description of your purported data and the approach you have planned to take. That's good. However, I still feel the language used is verbose: you can trim down in a shorter and more comprehensible language.

$\bullet$ The main meat of the post consists of the questions, which, albeit are related, are simply too many for a single post. This is not the norm here. Perhaps you can make additional posts citing them together for them to be answerable.

$\bullet$ Finally, the last para. It simply adds nothing to the query of the post. It involves greetings and other things like the usage of meta CV. This is simply not needed. (See this Meta post for additional details.)

Crux of my criticism is while there is definitely a question that could be entertained by the community and provide useful advice, it is certainly camouflaged among irrelevant statements: come straight to the technical content without beating the bush. Remember any potential answerer wouldn't waste their time to read a long post to come to a conclusion. Note the length doesn't matter unless it delves into a broader confabulation.

I appreciate you raised it here and would be happy to vote to reopen if you carefully exclude those unrelated statements and shorten your query to one or two questions.

  • 1
    $\begingroup$ Since I don't specialize in the field the question is based on, I have refrained myself from commenting particularly on the statistical relevance of the questions. The whole commentary is a general take. $\endgroup$ Commented May 19, 2023 at 4:26
  • $\begingroup$ Thank you, I really appreciate the feedback. In general I talk too much :-) I will take these suggestions to heart and try to come up with better questions. Much appreciated for your honesty. $\endgroup$
    – pandora
    Commented Jun 4, 2023 at 17:20

The first answer here by @User1865345 is excellent, and I agree with it all. It has the merit that it applies not just to the specific question originally raised by the OP. but also to many others.

Here is another take that is also very general. It's hard for anyone new to a field to imagine that they are answering their own question -- they wouldn't be asking if they could! -- but you can try some role reversal. One way to do that is to think of any topic on which you are well informed, something you've studied in depth, a sport or hobby that is a major part of your life, your views on politics, or your home town. All of a sudden an alien speaking English appears asking several loosely linked questions on that topic. In your role reversal, consider

How long would the ideal answer to those linked questions be?

How long would it take even an expert to write it?

In turn, here are my personal takes on those simple but crucial points.

If I answer a question, it is necessary that I think I can answer it with a few minutes' work. I have to think that, even if I am fooling myself, and in practice I will spend more because I enjoy answering, think I have something to say, and work on giving a good balance of principle and example. But if I think there are too many questions being asked, I back off and vote to close even if I think I understand the questions. People answering have to think that it is possible to write a good reply fairly quickly.

Comments on how long an answer might be boil down to the same point. If you seem to be asking for the equivalent of a review paper or book chapter, it is unlikely that anyone wants to spend time posting that on SE.

There aren't absolute rules on this. Sometimes people do write impressively authoritative, comprehensive and detailed answers, but that is less likely if a thread starts with a series of loosely linked questions.

  • $\begingroup$ Same as below, thank you for your honest feedback and I will try to listen and write better questions. $\endgroup$
    – pandora
    Commented Jun 4, 2023 at 17:20

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .