2
$\begingroup$

The mission statement of this site includes the sentence "With your help, we are working together to build a library of detailed answers to every question about statistics."

In particular, there seems to be a problem with newbie question treatment. Often newbies lack the facility to pose a question properly, and that question is often crouched in terminology that is somewhat self-contradictory, a not unexpected consequence of not knowing enough stats to pose a question properly. Sometimes those questions are closed when there would are simple fixes that could salvage those questions. Similarly, sometimes answers are downvoted because of inaccurate language, even when those questions grossly correct .

Is there some way we could better reward editing of question/answers to fulfill the site mandate than to discourage participation by eliminating both questions and answers? One is mindful that there are more questions than answers in general and a lot of sloppiness in everyone's linguistic rendition of both. Some answers are so complicated that they are inaccessible to some readers, some answers are so simple that they are truth bending.

It does not seem that we are heading in the direction of building up a reference guide as efficiently as possible. In particular, if we can reduce the negative treatment of both questions and answers and enhance the fixing-up of both we might increase that efficiency, as well as create a more wholesome experience for all participants.

Are we doing our best to make newbies, moderators, and everyone in between feel that their contributions are worthwhile? I do not think so. There seems to be a pervasive attitude of objecting to "stupid." There are a lot of dumb rocks in this man's universe, and if we try to kick them all we will just be wasting time. When we find a glimmer of intelligent thought, we should contribute to it as intelligent thought is a very rare occurrence.

Here is an example for @whuber. The original question, if memory serves, I rescued from the close vote stream, because it was interesting enough despite the improper use of the word "family" and the resulting "unclear" close votes, that I tried to answer it. The question is now +7/-3 (2018-08-26). Shortly thereafter, there were more downvotes than upvotes for my answer. However, my answer was eventually accepted by the OP, and has stood the test of time in that it now has twice as many upvotes as downvotes +6/-3. Even the confusion concerning what a "family" is merited a Q/A of its own that both @whuber and I tried to contribute to. That question is currently +11/-1. Whuber's answer is +9/-0, accepted by that OP (me), and so complicated that I provided an answer, +1/0, myself that was as simple as I could make it.

Now, without squinting at the original question to see the potential value in it, the entire chain of events would not have transpired. In summary, trying to extract value in an idea goes beyond the superficial form of the question itself, and without that effort to polish the apple, so to speak, many upvotes and a few downvotes would not have occurred. I think that the apple polishing should be enhanced and we should consider some type of reward structure for "apple polishing", not because I am seeking a reward, but because it goes toward completion of our mandate. Let's face it, there are lots of concepts that are difficult to grasp, and it is precisely those difficult concepts that are the most interesting ones. If the participants spend too much time kicking dumb statements and not enough time picking out diamonds in the rough, the overall experience with the site will remain not as rewarding as it could be.

Here is an example of a question currently in the close question queue marked as "unclear" three times and as "primarily opinion based" by myself. Now, what is unclear about the question? I do not think it is unclear, I just think it is speculation that is not particularly illuminating. What do you think? My comment was "I'm not sure that that is worth doing as one can just perform a minimization to obtain a best model." None of the categories seemed particularly appropriate. What do you think? Were we helpfully saying the question is "unclear." Can you understand it? My understanding did not seem to be especially taxed and it often is in the review queue. Were the other reviews helpful? What is so difficult to understand about this question?

Here is another question that was marked as unclear. However, I provided enough detail in my answer that it clarified the question and was taken off of the list of questions to be deleted. What this proves is that if one really wants to help, one can clarify in an answer what was unclear for both the OP and the reader.

From the above, there seems to be a single outstanding issue, and that is that "unclear" is often in the mind of the beholder. "Unclear" is subjective. So, let us put this as a clear question.

What can we do to mitigate inappropriate "unclear" closure votes?

$\endgroup$
11
  • 2
    $\begingroup$ Could you provide some examples of what you're describing? For instance, could you link to one or two questions that have "simple fixes" and explain what those fixes would be? $\endgroup$
    – whuber Mod
    Commented Aug 26, 2018 at 19:35
  • $\begingroup$ @whuber I could, but, I do not want to cast aspersion in anyone's personal direction. I have to great a respect for the participants to do that. Please consider that a call to accentuate the positive experience of participants is vanilla enough that it does not really require a formal proof of concept to be well understood. If you insist, I can rummage around for one, but I also spend some time putting stuff out of the close vote stream, so there are lots of occurrences. $\endgroup$
    – Carl
    Commented Aug 26, 2018 at 19:44
  • 2
    $\begingroup$ Without such an example, I'm unable to understand what you are describing. Perhaps you could invent one or paraphrase one anonymously? $\endgroup$
    – whuber Mod
    Commented Aug 26, 2018 at 19:49
  • $\begingroup$ @whuber OK, next one I come across I will document. $\endgroup$
    – Carl
    Commented Aug 26, 2018 at 20:06
  • 2
    $\begingroup$ "There seems to be a pervasive attitude of objecting to "stupid."" << That's a good thing, right? $\endgroup$
    – usεr11852
    Commented Aug 27, 2018 at 0:14
  • 2
    $\begingroup$ @usεr11852 I once upon a boy who was crying then started laughing. He then bent over and charged head first into a brick wall, knocked himself out, came to, started crying, then laughing again, and headed for the brick wall. I asked, "Why?" "Because it feels so good when the pain stops," said the boy. Well, knock yourself out, so to speak, objecting to stupid can be a lifetime occupation, just not a very rewarding one, I venture. $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 1:01
  • $\begingroup$ Still struggling with "grossly correct because of inaccurate language". "grossly" is a word I would only use negatively. Do you mean "grossly incorrect" or "broadly correct despite inaccurate language". $\endgroup$
    – Nick Cox
    Commented Aug 28, 2018 at 12:59
  • $\begingroup$ @NickCox Changed word order to mitigate the ambiguity. It was supposed to be read "grossly correct, (but) because of..." and I agree that was not clear. It is also very funny because the sentence suffered from the same language problem is was describing. BTW, I would like you to get two points for editing something like that. $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 15:09
  • $\begingroup$ Carl I appreciate this question and the anecdote about the boy knocking his head into the wall. :) I also agree with @whuber about the need for examples. $\endgroup$
    – Alexis
    Commented Aug 30, 2018 at 19:34
  • $\begingroup$ @Alexis Serious suggestion, do some reviews on the close vote queue and pay attention to what is at play. I seriously think that some of the reviewers find the questions "too dumb for words" or "trivial and uninteresting" or "hopelessly mixed up" and yet have no checkbox to say so. I really think we should explore what our motives are for closing things, not so much as to find fault, just to find a better fit for what we actually are doing. $\endgroup$
    – Carl
    Commented Aug 30, 2018 at 20:36
  • 1
    $\begingroup$ "Serious suggestion, do some reviews on the close vote queue and pay attention to what is at play." Uh... I do precisely that, but I agree with the introspection. $\endgroup$
    – Alexis
    Commented Aug 30, 2018 at 20:58

5 Answers 5

Reset to default
11
$\begingroup$

The problem with trying to edit newbie questions is that they are sometimes so incoherent that editing would have very real risk of making them something other than what the OP wanted.

I often link to a blog post I wrote called "How to ask a statistics question"; that has what I hope are some helpful pointers.

The question that was linked to is certainly not what I would consider a "newbie" question.

$\endgroup$
5
  • $\begingroup$ The question was posed On Dec 13, 2017, when the OP, @HardCore, had a reputation of 15. HardCore's experience since then has been positive enough to become a contributor with a current reputation of 466. Would that have happened if HardCore's early experience had not been overall positive? $\endgroup$
    – Carl
    Commented Aug 26, 2018 at 22:45
  • $\begingroup$ Agreed, but irrelevant. For example, unless I know multiple fold more than what is revealed in a garbled text, in that particular subject area, I tend to skip offering a close/open opinion. The alternative is to close everything I do not understand as "unclear" which is not a fair choice, and it is precisely those that I either spend considerable time trying to understand, if and when I am curious about them, or I skip entirely. $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 14:50
  • $\begingroup$ It is currently too easy to just say that something is unclear and too frequently I find myself understanding questions that others dismiss for some minor problems that are unrelated to the meaning of what is being asked. The close vote queue seems to be frequently the longest one needing attention. We users need some positive reinforcement to pay proper attention to it. $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 14:54
  • 3
    $\begingroup$ Actually, for me, it's just the reverse! If I look at the close votes for being unclear, it's often the case that the person voting to close understands the problem better than I do and that I missed some subtlety that they caught. $\endgroup$
    – Peter Flom Mod
    Commented Aug 28, 2018 at 9:50
  • $\begingroup$ Both occur. That is, sometimes "unclear" is used appropriately, sometimes not. When used appropriately, it is not problematic, when used incorrectly it is confusing and "off topic" so to speak, thus problematic. $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 18:26
1
$\begingroup$

One idea for a change which could make it easier to give a comment with questions/proposals to the OP, and not VTC immediately: When a comment is left and nothing happens in, say, 30 days, I am made aware of that fact, and can only then VTC.

$\endgroup$
2
  • 4
    $\begingroup$ I don't agree with this. In practice, 30 days is too long to wait. In practice and in principle, a comment can be anything from a good approximation to an answer to a request for clarification, so automating responses is not the solution here. We have to read the question and whatever comments and answers have appeared so far to judge what might else be said and what voting or voting to close is appropriate. $\endgroup$
    – Nick Cox
    Commented Aug 30, 2018 at 14:45
  • 7
    $\begingroup$ I think that it's acceptable to ask a clarifying question and VTC at the same time. If OP clarifies before it's closed, you can retract your vote. If OP clarifies after it's closed, the question can be re-opened. $\endgroup$
    – Sycorax Mod
    Commented Aug 30, 2018 at 15:54
0
$\begingroup$

There have only been 579 people ever awarded an explainer badge, i.e., people who edited and answered a single question, only 26 people who have done that 50 times; the refiner silver badge, and only two who have done that 500 times; the illuminator gold badge.

How do you feel about awarding +2 reputation for each time someone edits and answers a question? That might stimulate people toward pulling interesting stuff out of the close vote cue.

$\endgroup$
16
  • 2
    $\begingroup$ It's prone to be abused, just like the badge itself. Self-disclosure: I somewhat abuse this on another site by editing (even only copyediting) the question after I answer it to get the badge. Of course, I also edited and salvaged some questions that I don't know the answer. $\endgroup$
    – Andrew T.
    Commented Aug 27, 2018 at 9:52
  • 3
    $\begingroup$ That would be an SE-wide change that would require the developers to implement. If that is what you're after, this should be raised on meta.SE, not meta.CV. I doubt that this would be adopted, though. $\endgroup$
    – gung - Reinstate Monica Mod
    Commented Aug 27, 2018 at 13:42
  • $\begingroup$ @Gung Without some request coming from our site, the chances of a system wide change or a local sire change occurring is smaller than it would otherwise be. You have recently witnessed the uber-left cure all non-solution to excess negativity the current site structure is producing, the politically correct new "be nice" policy. Part of this is due to downvoting problems that show up as negative comments with some individuals calling others "stupid" out of small mindedness born of frustration. Nice language does not address those... $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 14:07
  • $\begingroup$ @gung ...con't problems, it just covers them up. I lived behind the iron curtain for 6 years, I do not recommend turning in your neighbors for perceived political offenses, trust me, it is the road to hell. We need a mechanical solution to excess negativity, and it will be increasingly problematic unless addressed. That is what is in play here. Only positive reinforcement can address the social degeneracy that population pressure creates. If you do not like my suggestion, come up with something else, please. $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 14:14
  • $\begingroup$ @AndrewT. Someone getting a badge or a few points that they cheated to get does not really hurt anyone, and there are some safeguards in place already. Problems with positive reinforcement from stealing cookies from an infinitely large and imaginary cookie jar can be addressed, but that "theft" would not hurt others like negative personal attacks do. Here is the cost benefit; the cost is imaginary, the benefit is real. $\endgroup$
    – Carl
    Commented Aug 27, 2018 at 14:40
  • 1
    $\begingroup$ Presently, there is an automatic award of +2 rep for editing, but only for users whose total rep is <2000. Given that this edit-only award has an even lower bar that you propose (i.e. edit + answer) and the problem still exists, would your proposed solution change anything? For high rep users, a +2 bonus is unlikely to influence motivation. And for low-rep users who are willing to edit a question, they would probably answer it (and gain more rep by upvotes) if they could. $\endgroup$
    – mkt
    Commented Aug 28, 2018 at 9:47
  • 2
    $\begingroup$ (Is this offered as the answer? You are clearly allowed to answer your own question but this proposal seems like a small detail compared with the important and difficult issues you raise.) FWIW, I doubt that anyone with a reputation much over say 1000 is all motivated by getting an extra +2 here or there, but I speak only for myself. $\endgroup$
    – Nick Cox
    Commented Aug 28, 2018 at 13:02
  • $\begingroup$ @NickCox On meta there are no points for anything, and yet we participate. A positive reinforcement would have to fit into the site's style fairly seamlessly. Is there a proposal that you would like better? $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 14:55
  • $\begingroup$ @mkt Understood. How about +4 for editing and answering but only for users whose reputation is >2000? $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 14:58
  • 2
    $\begingroup$ I have contributed to several threads with this theme but regret to have no new ideas. It's unfortunate that many new or occasional posters here don't realise that CV (or SE) isn't a help line. It's understandable that many people who use statistics understand it poorly. The net outcome is that many questions here aren't even answerable, let alone contributions to the long-term improvement of the site. I have no objection to anyone who wants to be nice, gentle, kind -- I try it myself much of the time -- but I fear what the site needs most is ruthless weeding out of questions that don't fit. $\endgroup$
    – Nick Cox
    Commented Aug 28, 2018 at 15:06
  • $\begingroup$ @Carl I wouldn't object to something of that sort - such incentives can't hurt and I'm not too worried about excessive/trivial editing. I tend to agree with Nick Cox's view in the above comment, though. $\endgroup$
    – mkt
    Commented Aug 28, 2018 at 15:56
  • 1
    $\begingroup$ @mkt So why do we not expand the definition of "low quality" enough to cover the low quality posts? Currently "Very low quality (i.e. no amount of editing can salvage the post) (only new posts scoring 0 or less)" Right now, we are throwing out stuff that has value, mislabeling low quality as "too broad" or "unclear." $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 17:08
  • 1
    $\begingroup$ @Carl Fair point. I tend to think that the language around flag/close reasons is too narrow. I've slowly learned that the community doesn't it fact stick to the text strictly in practice. I think this is a good thing, on balance, though fixing the text would be better. $\endgroup$
    – mkt
    Commented Aug 28, 2018 at 17:12
  • $\begingroup$ @mkt Agreed, but the worse effect is upon the OP whose post is inexplicable mischaracterized which misdirects the OP's corrective action. Also, we still need more positive reinforcement. $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 17:26
  • $\begingroup$ @NickCox I do not know how to distinguish between reasons for turning to this site with a question that are considered help or not help. Certainly, some questions do treat us like a help line, e.g., this or this, but do we actually have a mandate to draw a line, and if so, how? $\endgroup$
    – Carl
    Commented Aug 31, 2018 at 5:18
0
$\begingroup$

Expand the definition of "low quality" enough to cover the low quality posts? Currently "Very low quality (i.e. no amount of editing can salvage the post) (only new posts scoring 0 or less)" Right now, we are throwing out stuff that has value, mislabeling low quality as "too broad" or "unclear." The worse effect is upon the OP whose post is inexplicable mischaracterized which misdirects the OP's corrective action.

$\endgroup$
13
  • 1
    $\begingroup$ The boundaries here are often (hmm) unclear. Too broad, unclear, low quality aren't mutually exclusive: two or three such labels often make sense. We've had posts where the votes to close were all or almost all for different reasons. I don't get that rejecting for the wrong reason is much of a problem even in abstraction. We are, supposedly, "mislabeling" or "mischaracterizing". What does this mean except that in some cases other people select different criteria from you? Where's the objective evidence? $\endgroup$
    – Nick Cox
    Commented Aug 28, 2018 at 18:06
  • $\begingroup$ @NickCox Well, you just gave that evidence. One could collect it objectively, I suppose. On some questions, the agreement for closure is uniform, all choosing the same reason for closure, on others, there are three or more reasons, none of them really saying what is going on. Sure, there can be objectively more than one reason for closure, however, it also may be that none of the reasons is a good fit. $\endgroup$
    – Carl
    Commented Aug 28, 2018 at 18:18
  • $\begingroup$ There is plenty of evidence that people disagree on reasons for vote to close. Some people might want to interpret that as that some of the reasons are wrongly declared. $\endgroup$
    – Nick Cox
    Commented Aug 29, 2018 at 7:37
  • 1
    $\begingroup$ @NickCox When a multiple choice has no good answer, the answers chosen tend not to agree. If we improve the choices, the editing and OP response to it would be more directed. $\endgroup$
    – Carl
    Commented Aug 29, 2018 at 15:27
  • 5
    $\begingroup$ In principle we could discuss different wording, but I think it would be tinkering with small details that don't matter much. In practice if a question appears poor on several different grounds it is almost always doomed any way. I don't have very much time or inclination to try to rescue really bad questions. More positively, there are enough mechanisms for those willing to try: comments as specific and as constructive as possible and directed at the question are immensely more important than precisely which reason is given to the OP for why a question was closed. $\endgroup$
    – Nick Cox
    Commented Aug 29, 2018 at 16:11
  • 1
    $\begingroup$ I rarely disagree about reasons for VTC. The problem I run into is that many questions ought to be closed for two, three, or more of the reasons available and I can pick only one of them! I vote for the one I think might give the most constructive suggestion. Carl, I haven't figured out what you mean by "throwing out stuff that has value." If a question is not in a shape to be answered and the OP doesn't improve it, it's difficult to hold to the position that it's somehow still "valuable" to the site. (cc @Nick) $\endgroup$
    – whuber Mod
    Commented Aug 30, 2018 at 20:37
  • $\begingroup$ @whuber I am not concerned here about those questions that are "so incoherent that editing would have very real risk of making them something other than what the OP wanted." That is not the topic that I have been trying to broach. That is the most popular answer, but it is not germane to my intent, and it is not a problem. $\endgroup$
    – Carl
    Commented Aug 30, 2018 at 20:59
  • 2
    $\begingroup$ It's unclear to me, then, what you are concerned about. You explicitly posit a situation in which a question is "self-contradictory." Although you refer to "simple fixes" you don't appear to explain what you have in mind. That quotation you draw from Peter Flom's answer tried to point out that the obvious "simple fix"--namely, editing the question--won't usually work in those cases. The fact that you're getting answers and comments all over the place indicates your question here on meta needs work. Could you focus and clarify it? $\endgroup$
    – whuber Mod
    Commented Aug 30, 2018 at 21:17
  • $\begingroup$ @whuber I am trying to solve the following problems. 1) The close queue is too long and reviewing it should be rewarding for those doing the work, or it will stay too long. 2) Some questions are valuable, like the one referenced, and should be saved by editing or just answering them. 3) We need better positive reinforcement and rewards for our work and good deeds. We know this because the new policy for "niceness" is a reaction to surfeit negativity. It is a reactionary policy, not a proactive one. I want proactive activity. Finally, these three things are related. $\endgroup$
    – Carl
    Commented Aug 30, 2018 at 21:43
  • $\begingroup$ @whuber I am asking here for ideas people find acceptable to foster positive reinforcement. That some questions should be closed and are unsalvageable is beside the point. That if we want to be negative, we should do so politely is beside the point. The question goes to how can we be more positive, not less negative. $\endgroup$
    – Carl
    Commented Aug 30, 2018 at 22:12
  • $\begingroup$ @whuber I am trying to explain. See new text above, another example. Do correct me if I am wrong, but, I think I understand the question better than the responses. $\endgroup$
    – Carl
    Commented Aug 31, 2018 at 4:50
  • 3
    $\begingroup$ That doesn't surprise me, because in these comments alone you have articulated at least three distinct questions. That would explain why your edits are getting so long that the post is becoming unreadable. Please simplify it to focus on one question. $\endgroup$
    – whuber Mod
    Commented Aug 31, 2018 at 13:32
  • $\begingroup$ @whuber Fine, I boiled it down to a single question. $\endgroup$
    – Carl
    Commented Sep 3, 2018 at 19:49
0
$\begingroup$

We could award +1 for users with reputation >2000, who leave a first comment when they vote in the close vote review queue. That might help the OP to understand why their question is on the chopping block, and take remedial action, and also, it might clear up the close vote queue. If we do not do something like that, the close vote queue will continue to be blocked. Taking undue advantage of leaving comments would not likely be abused because the comments are there for all to see.

$\endgroup$

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