As I have researched and tried to learn how SE works by reading different comments, answers, questions and blog posts, I have come to the conclusion that SE is interested in becoming the Q/A site providing the highest quality Q/A pairs.

For this to be possible, there needs to be some way for the content creators to easily see how they can improve their content based on some easily digestible data (feedback) provided to them.

My first thought was that the voting functionality is the data I should trust on to decide if my question or answer is of high enough quality. But after researching the voting functionality and its underlying use-case this seems not to be the case. I have received feedback suggesting that the voting functionality is mainly there to improve visibility of content which already is considered to have a high enough level of quality.

The voting functionality also has multiple different meanings and on top of that is different between Meta and main sites. So I cannot know which meaning the downvote actually has.

Then I thought about comments. Maybe feedback on how to improve my content would be provided via this channel. This actually wouldn’t be any measurable evidence, as words are based on opinions of the writer. Also, by further research, I have come to the conclusion that providing feedback on how to improve the content is not seen as necessary here.

One answer is suggesting the Question Wizard. While i see the intend and usefulness of the tool to improve the initial Question. It is not helping to improve the Quality of the Question, not to mention the answer, after it was created. Furthermore it is not a metric of quality, it merely is a tool to help me understand what the community perceives to be a question of high enough quality.

It also suggests that after submission i can only "Hope for the best". I believe that this is not a good approach at all. If i just hope and act randomly without any measureable evidence no quality is achieved. You can see that if you observe AI Systems which are provided with to less Data and Feedback. They act randomly but never evolve in any meaningful way.

As those are the only two metrics I can see in an obvious way, I now ask myself: On which data should I base my actions on how to improve my content?

  • 1
    "This actually wouldn’t be any measurable evidence, as words are based on opinions of the writer" Could you give an example of feedback where this is not the case? Even off site or theoretical. You could say the same voting, viewing etc. How could feedback not be someone else's opinion? – dustytrash Mar 10 at 3:35
  • To correctly answer you question: No, there exists no Feedback Model which is not based on someone's opinion. – Kevin Kötz Mar 10 at 7:28
  • However, there exists a feedback model which immensely reduces the propability for Missinterpretation. If you want to learn about it, I suggest this publication: researchgate.net/publication/… – Kevin Kötz Mar 10 at 7:29
  • It is based on the Idea that a simple yes/no to a question can not be misinterpreted. Only the question itself could. – Kevin Kötz Mar 10 at 7:31
  • To reduce the propability to have a misinterpretation there, you will need to have it very easy. – Kevin Kötz Mar 10 at 7:33
  • To prove by example: "Is my chain of thought clearly understandable for you?" Or "Does this question need improvement in it's readability?" – Kevin Kötz Mar 10 at 7:35
  • The second question can be transferred to content on SE. If SE sees the readability as it's main problem they could introduce a "Readability Metric" which is presented to the content creators. – Kevin Kötz Mar 10 at 7:37
  • From professional experience this metric needs to be percentage based. As percentages of something are well understood (or felt like it) by the general public. – Kevin Kötz Mar 10 at 7:39
  • So if SE would introduce a Metric about Readability and present it to the content creater as "This amount of users think your content is hard to read: xx%" The content creators would intuitively know what to do. – Kevin Kötz Mar 10 at 7:41
  • If this metric then is linked to a easily understandable and well written FAQ, the content creators would have a way to better understand what readability means. – Kevin Kötz Mar 10 at 7:53

Welcome to the real world.

Basically the only thing you missed in your list is the new "question wizard" that was introduced (more or less) recently and that aims at providing some very basic feedback while you are writing the question. See The new ask page is now live on the network!

After you submitted it, you can only hope for the best. Downvotes can indicate that readers have issues with your content. If so, you can hope that one or more users leave a helpful comment. And of course, when enough close votes come together, you get to see the close message. Which you can then research in depth.

Worst case, you come back, and your question has even be delete voted and is gone.

And besides turning to a meta community asking "what is/was wrong about my question X" there isn't much you can do.

As you can see, nothing in the whole process (maybe besides the very last option about meta) will necessarily lead to measurable evidence based feedback. That can happen, but there is no way to enforce it from the perspective of the questioner.

|improve this answer|||||
  • I was in reality the whole time :) It's just my very own one. All my thoughts and questions are coming from a psychological and philosophical point of view. – Kevin Kötz Mar 10 at 7:43
  • The misunderstanding happening here was/is coming from the concept that two people with completely different backgrounds tend to miss understand each other more easily when talking about Topics in the Area of Expertise of any of them. – Kevin Kötz Mar 10 at 7:47
  • I currently have no evidence to prove it to you. But please trust me, the statement has a very high degree of trueness to it. – Kevin Kötz Mar 10 at 7:48
  • I recommend reading my comments on the original answer to understand the intent of the question asked. – Kevin Kötz Mar 10 at 7:49
  • 3
    Well, I answered your question: there is no such thing you are looking for. As of now, at least. If you have good ideas how to change that, then you might rather write up a feature request. – GhostCat salutes Monica C. Mar 10 at 7:53
  • You did not answer the Question. The Question is about measureable metrics/data/evidence. Can you explain how the question wizard or Hope is a measureable data? I edited my question according to the feedback gathered by your answer. – Kevin Kötz Mar 10 at 15:16
  • 2
    The answer is: that doesn't exist, at least not in ways that you can enforce to manifest. I explained the "available status quo" to you. And note: please be careful about editing questions like that. The basic idea here is to use editing to improve quality by fixing unclear wording, etc ... the idea is not to turn questions into moving targets. You asked "how does X look around here", and the answer is "X per se doesn't exist here". – GhostCat salutes Monica C. Mar 10 at 15:47

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