How can we encourage new users to edit their closed posts?

Question

New or inexperienced users are sometimes faced with their question (or answer) being closed without any detailed feedback, or even without any comments on their post.

While this is definitely the way to handle spam, it would frustrate users in particular when they put quite some effort into their post. I am afraid that it would even scare them away from CV.

⮚ shouldn't serious users get least one individual feedback before their post is closed?

Here is a recent example. The user split up their question into four points, addressing various thoughts they had when trying to analyse their data. It thus became a very long and complex question that is probably too complex to answer as a whole. Therefore maybe no answer can be accepted as "answering the question to the fullest". However it is clear (at least to me) that this was a thoughtful question, but for the user the answer is: Closed. This question needs to be more focused. It is not currently accepting answers. A decision was taken within three hours after the question was posted, but apparently no discussion in comments going on before vtc.

I am afraid that this can make users think they are unwelcome, and that no-one bothered to actually read their question.

What does the community think about this issue?

Edit:

Personal comments may lead to an endless discussion about closing reasons, and it is perfectly understandable if they are avoided.

Would it be an option to have automatic replies for new users that indicate that it was not a machine that closed their question?

Such as

Welcome to CV. At least three reviewers have read your question and found it needs to be more focused. It is therefore not currently accepting answers, but it will be reconsidered when edits have been made to improve it.

Answer 1

I don't think there can be a simple yes/no answer to it.

As I am one of the close voters in the specific post mentioned, here is my two cents:

I am of the conviction that the message

Closed. This question needs to be more focused. It is not currently accepting answers.

which is followed suit by this:

Update the question so it focuses on one problem only. This will help others answer the question. You can edit the question.

couldn't be clearer in conveying what the reviewers were thinking about the post in concern. A vtc is not at all the end of the journey, rather, I believe, it is one of the genuine steps to help OP - any positive feedback in tune to ameliorate the post would increase the chance for getting a valuable answer in return.

I, myself, routinely try to leave a welcome message that might also be mentioning any additional thing(s) OP needs to address. But whenever I feel the default closure reason is enough, I don't add any extra comment. I respect each reviewer who volunteer to clear the queue by taking out some time from their schedule. If they refrain from any customized comment, that is not tantamount to any form of rudeness or unwelcome attitude towards OP.

If anyone wants to add any comment airing the same thing as the default closure statement, they can do that readily. But I feel it is redundant. And absence of such isn't akin to any form of reprehensible tone, I reiterate.

Response to the edited query:

Would it be an option to have automatic replies for new users that indicate that it was not a machine that closed their question?

Is this direly necessary? If you notice, there is a link in the closure message that leads to this statement (emphasis mine):

Questions that need additional work or that are not a good fit for this site may be closed by experienced community members. Closed questions cannot be answered, but can be edited to make them eligible for reopening.

There is a detailed elaboration and each point has been properly articulated for any new user to comprehend the situation. We expect from any new user a minimal attitude to at least immerse in a bit to check the norms here.

If you really want to add something softer or more humane touch by any standard, you are welcome to do so. But as gung - Reinstate Monica indicated, you have to keep in mind the sheer load of the queues to be reviewed - only the mods and few users bother to clear up the pending queue; we appreciate new users of the community to come forward and volunteer in the same vein.

IMO, the default message really works well and so, no additional message is explicitly needed to articulate the same thing.

Answer 2

My policy is that I have four tiers of responses:

1. If it seems that OP didn't put minimal effort to write the question like properly formatting it, doing a trivial Google search, reading it before posting to notice that it does not make logical sense, or checking that the question would be bluntly off-topic, I just vote to close without comment. If the person asking doesn't care, why should I?

2. If they put in minimal effort, but still the question is an obvious candidate to put on hold, I vote to close and comment.

3. If they put some effort into asking and the question is fresh, I ask for clarification without voting to close. I try to re-visit and if there was no reaction after several hours, then I vote.

4. If I feel that OP did put some effort and the question is an edge case, sometimes I try answering it even if it's slightly unclear, near duplicate, or a bit vague. In some cases it works, in others I wonder if I should stop doing this because I learn that the question was unclear enough that I misunderstood it, etc and my answer didn't help at all.

Answer 3

I guess, or hope, we all agree that it is better to be nice not nasty and that we want users to feel welcome rather than the opposite. Existing policy already allows detailed and friendly comment from those with time and inclination to give it. I often try to elicit better or more precise questions through comments, with mixed success.

It's encouraging that really rude or offensive comments almost never appear on CV and are edited out swiftly if they do. (The rudest comments I've seen have been directed at moderators or long-term supporters.) The standard of civility and helpfulness here vastly exceeds what can be seen e.g. on Twitter, let alone worse places, and I don't think we should beat ourselves up as being unfriendly.

But there are other statements of equal importance here.

Cross Validated specifically and Stack Exchange (SE) generally do not offer help lines. On SE there is no entitlement to an answer. People must ask a good question first, and even then no-one is obliged to answer. I consider that CV is a good thing and would not be active here otherwise, but a problem for CV -- for new users and experienced users alike -- is that many threads are of very limited value. There are many reasons for this, but in total it is why people should maintain, if not strengthen, the existing barriers whereby, given sufficient reputation, you can vote to close, or downvote, or explain why a question is not yet a good question, in your view.

SE is well documented. I consider that new users who are disappointed are partly to blame for their disappointment if they don't read around to discover expectations or limitations which imply why their question hasn't been accepted. It's not really a good use of anyone's time to write out specific messages explaining why (for example) a software-specific question is off-topic here -- or a request for advice on how to analyse someone's data is impossible to answer well without much more detail on their goals and indeed their data. I have done my share of that, but I don't feel personal or collective obligation to do much more.

As often pointed out, statistics is a field where many people briefly have an obligation to study or apply statistics, but they do not have a long-term incentive or goal to put much more effort into learning it thoroughly. That is often fine but it also leads to many of the more naive posts which boil down to requests to do the posters' thinking for them.

Answer 4

Stack Exchange jargon often uses common English words but with different meanings.
New users, and sometimes even we older ones, are often confused by situations like:

• If one asks something about X on X.SE, how can it possibly be "off topic"?
• If one quotes some authenticated facts and asks a question about them, how can the question be "opinion based"?
• And what does "focused" mean when the question doesn't even include an image?

Whenever I vote to close (and sometimes for already closed questions), unless someone has already given a good explanation, I usually add a comment explaining why it was closed.
In particular, I point out specifically what was wrong and how the question could be changed to correct the problem:

• The question asked is different from the summary question in the Title.
• Three related questions should be posted separately.
• Hearsay is posted as well-known fact where quotations and citations are required.
• It's much too long with far too many irrelevant details.
• It violates a restriction (specifically describe it) that is listed in the site's tour (include link).
• Etc.

For instance, just yesterday, a post on Skeptics.SE was looking for examples of claims of spinal-cord faith healing, presumably because the OP was skeptical that such claims were real. The question certainly looked reasonable for a "skeptics" site, but it was closed for being "off-topic".

So I added this comment:

To be an acceptable question for this site, it must contain a quotation either from a well-known person or from a well-known publication. In this case, it would have to be a quotation that explicitly reports a specific case of spinal-cord damage being healed through faith healing. Without such a quotation, there isn't any notable claim here to be skeptical about. (And yes, I know you are trying to find such claims, but that isn't what this site is for.)
— religion - Has faith healing ever repaired spinal-cord damage? - Skeptics Stack Exchange

There is a very steep learning curve on Stack Exchange sites, and I'm sure many new users are turned off by this attitude and give up.

Answer 5

This is also a question that has concerned me, at least sufficiently that I try to put a comment explaining my suggested next step — especially if I am the final person to vote to close a question, and the question seems in good faith but where the poster is likely unfamiliar with SE procedures in general or CV in particular. My recommendation is often to edit and refocus if I feel the question can be salvaged, or to ask the question on another site if an alternative SE would be a better fit (we can migrate to some other sites on the network as part of the closing vote, but not all). Quite often there's also an X/Y problem and my suggestion is to have a serious rethink before editing or re-asking. I might point in the direction of some similar questions which I think might be relevant, and may even solve the underlying problem, without being duplicates of what was actually asked.

As the above paragraph might suggest, I would like to somewhat challenge the framing of this Meta CV question. Sometimes we don't, in fact, want to encourage new users to edit their closed posts. A case that I'm never sure how to best deal with is when a poster asks a question that's both close-worthy from our point of view, and mis-specified from their point of view (they haven't asked the question they really meant or needed to ask, due to a faulty assumption or the X/Y problem). By the time the question's closed, it may already have answers, which will become irrelevant or confusing to future readers if the question is substantially edited and reopened. This highlights how it's important to close poor questions early, before answers come in — and one of the drawbacks of expecting reviewers to provide substantial personalised feedback is that this slows down reviewing, while discouraging experienced users with 5 minutes to spare from checking out the review queue.

On balance, I share many of the concerns raised by this Meta post. I have seen salvageable and potentially interesting questions that just fizzle out after closure without reviewer feedback. I don't want to blame reviewers in any way for not leaving feedback — it's unpaid work after all (free labour granted to a for-profit corporation, no less!), certainly the OP has no "right" to feedback just as they have no "right" to an answer, and the reasons that appear on the close notice are usually informative enough. Nevertheless, OPs seldom seem to act on them in the absence of feedback. Sometimes this will be due to "drive-by" question-askers, who may not have acted on extra feedback anyway. One factor in reviewers deciding how much effort to give in terms of feedback and encouragement, is their perceived likelihood of the OP being drive-by: after years of reviewing you do get a decent sense of this. So you'd have to be careful about reading too much in to rates of OP editing and improving questions in cases with and without reviewer comments. Similarly, even if you restrict yourself to cases where OPs do make edits, I suspect you'd find the biggest quality improvements come in cases where reviewers leave useful and specific feedback that goes beyond the generic message on the close notice, but again reviewers will generally only leave such feedback in cases where substantial improvement seems possible!

With that in mind, the problem might be less severe than it first seems, but also rather hard to perform a valid statistical analysis on. Are there any actionable responses we could take anyway? I have a few suggestions intuitively, several of which might be suitable for A/B testing if someone at SE wanted a more data-driven approach. These are mostly technical suggestions because I think other answers have explained why you're unlikely to achieve the change you want via a policy or attempt at enforcing a new social norm in favour of reviewers giving more detailed feedback (and I say that as someone who often gives it).

Are there ways to make providing additional feedback easier for reviewers?

• We have a Meta list of canned responses to copy/paste that often provides more detail than the reasons given in the close notice. Could we promote this list more, so more reviewers are aware of it? (This is something CV might be able to do for itself, the other suggestions I think only work if there was buy-in from SE itself, which means they are really better-suited to SE Meta than CV Meta.)
• As an extension of the above, would reviewers benefit from drop-down commenting tools for common responses? This might be a useful addition to the Review Queue itself. I believe there are/were browser extensions for SE sites that serve a similar purpose.

Is the close notice too daunting or confusing for posters?

Generally they explain the reason for closure quite adequately, and they do suggest editing and improving so the question can be reopened. Nevertheless,

• we could try cutting down the verbiage so the notice is less intimidating,
• we could try rewording to emphasise the editing/improvement angle more clearly,
• we could try giving a less intimidating and less final-sounding name, e.g. "on hold" instead of "closed",
• even though it's basically duplicating info that's in the close notice already, we could try having a bot comment to suggest editing and improving. I like this suggestion enough to think it might be worth trying, since people sometimes do respond to redundant information, but it does come with risks e.g. OPs responding to the bot in the comment in the hope of drawing assistance but getting no reply.
• even more radically, should we distinguish between questions that look salvageable and those that aren't? There are some which fall so far below our quality standards, or so far outside our scope, that encouraging the OP to edit seems wasteful. I can see some benefits to having a checkbox in the close-voting menu to say the post looks salvageable, and the OP should receive feedback encouraging them to edit (whether that's via a reworded "on hold" notice rather than the normal "closed" — something which might be visible to the OP but not other readers on the site — or the bot comment suggestion, or some other method else entirely).