I came across this meta question asking about the quality of the answers on one of the questions on the main site. After looking through the answers to THIS question I also noticed most of the answers were not really..... answers. They were more guesses.

In my opinion, answers that start with 'I think' or, 'My guess is' should be flagged and removed, as this is supposed to be a Question and Answer site, not a 'Question and Guess' site.

I understand that this particular question would be difficult to give an answer to, however, if the 'I think' answers had provided sources, links, examples or other things along those lines to back up what they were guessing at, then those answers would be much more credible. I have seen this happen before, where people give answers that start with 'I think' and I always downvote them.

Should these answers be allowed to stay? Or should they be flagged and downvoted until they are able to provide sources to back themselves up? Because to me, as it stands, guesses are useless as answers.

I will point out that this question is not to do with this question and its answers in particular. It is in general across the whole site. If you agree or disagree, I would like to hear why.

  • \$\begingroup\$ Comments are not for extended discussion; this conversation has been moved to chat. \$\endgroup\$ – Nick Alexeev Aug 20 '18 at 20:08

I will point out that this question is not to do with this question and its answers in particular. It is in general across the whole site.

We shouldn't enforce a general rule against "I think" and "I guess", because it would be too Procrustean. Instead, we should ask for higher answer quality. Higher quality can be achieved by posting reference, or by providing one's own reasoning.

"I think" and "I guess" are disclaimers that the writer doesn't have authoritative sources. One can do much worse than say "I guess". One can post his guess without a disclaimer that it's a guess, and sound more authoritative than he should.

Because to me, as it stands, guesses are useless as answers.

As a general statement that doesn't hold. There are answers with explanations of rationale behind somebody's thinking. Those can be useful.


There is little reason to call for a mechanism not already baked in to the SE interface.

If the "I think" answers are wrong, downvote them and note why in a comment. There are reasons why answers are voted on, and this is one of them.

If the "I think" answers are right, comment to that effect, or, better yet edit the answer, as I don't believe comments are guaranteed to not go away.

If questions call for an opinion, unless there is some other merit to the question, vote to close it.

  • 2
    \$\begingroup\$ While "answers" laden with qualifiers to even their basic technical accuracy are improper, editing them to remove those qualifiers is even more improper, as it means fundamentally altering someone else's statement. Technically accurate answers with qualifiers as to their applicability are a little better, but tend to indicate a question that needs improvement. And there's a very real risk of long divergences into the irrelevant, particularly with less experienced askers who may not be able to quickly realize that one of these tangential rants is irrelevant. \$\endgroup\$ – Chris Stratton Aug 21 '18 at 14:14

"I think" answer can be useful. It can give the asker something to look into more. That's a lot better then not knowing what to look into at all. Maybe it won't work out, but it should be a lot more likely than looking into a random topic.

"I think" done right really means "I have some knowledge and experience with this topic, but don't know the answer definitively. Still, this off the top of my head answer is something that will probably point you in the right direction. I'm a volunteer here, and don't have the time or inclination right now to dig out the full answer. Remember that this could be wrong, but I wouldn't be writing this if I didn't think there was a good chance this can help." "I think" is rather more succinct and says the same thing.

  • \$\begingroup\$ Surely that's more of a comment than writing an answer. At least if I'm not sure about an answer I'll usually write it in a comment \$\endgroup\$ – MCG Aug 22 '18 at 6:28
  • \$\begingroup\$ @MCG If you're not sure about an answer you should definitely not write it as a comment, where it can't even be downvoted if it's wrong! \$\endgroup\$ – pipe Aug 22 '18 at 11:13
  • 1
    \$\begingroup\$ @pipe, I was speaking about the example Olin made. It's better having it as a comment in more of a 'hint' than a guess as an answer in my opinion \$\endgroup\$ – MCG Aug 22 '18 at 11:21

Questions that call for "I guess..." answers (because there's no way to know the correct answer), should be closed as "Opinion based".

For example, common questions in this category are

  • Why didn't [vendor] include [feature] in [chip]?
  • Why didn't [vendor] include [spec] in the datasheet for [chip]?

If a question doesn't require such answers, and somebody answers "I guess..." without some solid evidence or reasoning why it's a well-educated guess, then the answer should be downvoted.

  • 1
    \$\begingroup\$ if we close all questions which do not have a definitive answer as opinion based, we will close almost all questions. the remaining ones could probably be answered with "go and read the datasheet again" or similar. \$\endgroup\$ – Loganf Aug 20 '18 at 21:11
  • 1
    \$\begingroup\$ @Loganf, I've edited to give examples of the kind of questions I mean. We see these fairly often, and the real answer is "because they didn't think the resources needed to provide the feature or test the spec were worth the sales to be added by doing it." but we can't know the real reason without reading the minds of the people who made the decision. \$\endgroup\$ – The Photon Aug 20 '18 at 21:17
  • \$\begingroup\$ "Why didnt" questions tend to be a lot more guess-y than "Why did" questions. Remember, opinion based closing is for wild guesses, and do not apply to questions that can be answered based on common or best practices and industry standards. Why did x include this ic? Because that IC does z and that's standard in xyz designs. \$\endgroup\$ – Passerby Aug 24 '18 at 21:20

Yes, I think "I think" answers and the questions that require them should be allowed to stand if there is something on topic to be learned from them and they are written from a place of experience. If they dont stand up to scrutiny or the reasoning not well explained, they should be down voted.

Closing them as "opinion based" sends the wrong message because the question is then put in the same bracket as "which is better PIC or AVR?" or other polemic nonsense.

The linked question for example, cannot be answered definitively but does allow people with experience to provide their insights. It is possible that it's a normal practise which can be explained by someone who has done this first hand, but even educated guesses of the purpose of the capacitors are instructional. (Though it is a shame that the top answer is a collection of uninformed wild guesses from reddit, that is not the questions fault)

In my career, I've learnt far more from the opinions of senior engineers than factual answers they give. Most of the decisions I make in my work are opinions because getting objective answers would require prototyping and evaluating every single idea.

note that the community doesn't seem to agree with me, and the topic has been discussed repeatedly:

Is there/should there be a mechanism for asking opinion-based questions?

What about making a survey website

Is there a place for opinion on the electronics.stackexchange?

On "Opinion based answers"

  • \$\begingroup\$ My opinion is they should be downvote and flagged unless they provide something to back them up. Some links, examples or documents to support the theory. Take your answer on that question for example. It starts with 'this may be a crazy idea' and you go on to give some thoughts as to why your guess could be correct. Then you end with 'but I have never seen this before'. That to me is a blatant guess and is the reason for my downvote on it. There were no sources or examples. Those are the guesses I don't like. \$\endgroup\$ – MCG Aug 21 '18 at 10:33
  • \$\begingroup\$ Ones that have a few resources, examples, documents to back them, possibly some mathematical proof behind it (bear in mind I'm talking about all guess answers, not just ones on this question) should be considered good as it shows some research has been done and some effort put into the answer to show why it could be correct. As I said a few times, this site is supposed to be a collection of good questions and good answers. Not questions and guesses \$\endgroup\$ – MCG Aug 21 '18 at 10:37
  • \$\begingroup\$ @MCG - it is a blatant guess. no argument from me, though perhaps "crazy" was the wrong word to use in the intro. if you think my reasoning is wrong, please comment on it and explain why. what documents or resources or proofs would you like me to provide? which part of my answer do you disagree with? I'm not going to reference everything back to first principles, but i will put some effort in if you would like a specific clarification. (if the question is re-openned, which seems unlikely) \$\endgroup\$ – Loganf Aug 21 '18 at 14:09
  • 1
    \$\begingroup\$ I would suggest that rather than instinctively going "no sources = downvote + flag" we should be asking "does this answer make sense?", and "is it clearly reasoned?". in the case of that specific question "could this suggestion the best way of achieving the stated reason?" and "is it the sort of solution a professional engineer might come up with?" \$\endgroup\$ – Loganf Aug 21 '18 at 14:10
  • \$\begingroup\$ I've already explained why I downvoted the answer. I usually do comment on answers I downvote but there were I think 4 or 5 on that question hence deciding to do this rather than copy and paste the same thing under all of them. My comment would just have been "this is a guess with no resources" and as you said there it is a blatant guess, which was obvious, and hence my downvote reason needs no further explanation \$\endgroup\$ – MCG Aug 21 '18 at 14:14
  • 1
    \$\begingroup\$ Generally "catalog of possible reasons for failure" pages are not a fit for the Stack Exchange model, and so don't belong here. The possible exception is for stumbling block problems that beginners trip over endlessly, where having a canonical question to close the repeats as a duplicate of can be useful. But canonical questions are very hard to do right; there's hardly anything worse than what happens when an inexperienced contributor tries to artificially create one. \$\endgroup\$ – Chris Stratton Aug 21 '18 at 14:17

Let me repeat what I said in my original comment: If a person writing an answer is not sure about the question or his answer, he should either write a comment asking for clarification on the question, or do more research on the answer until he's confident in the information that he's providing. You're right -- "I guess" or "I think" answers are really not appropriate.

  • \$\begingroup\$ Appropriate or not, a new mechanism to get rid of them beyond community moderation is not called for. I don't see a reason to flag them, unless the "I think" is clearly wrong. \$\endgroup\$ – Scott Seidman Aug 24 '18 at 13:13
  • \$\begingroup\$ Also, "I think" might be a symptom of a participant. Sending a message from a diamond moderator saying "your post was deleted because you said 'I think'" is not in line with the new behavior guidelines. \$\endgroup\$ – Scott Seidman Aug 24 '18 at 13:35
  • \$\begingroup\$ @ScottSeidman you wouldn't get a message saying "your post was deleted because you said 'I think'", that's a bit silly. A message along the lines of 'this is a guess with no documentation or sources' would be more appropriate, and I think it would be more than fair. \$\endgroup\$ – MCG Aug 28 '18 at 11:02
  • \$\begingroup\$ @MCG is there a real reason such an answer couldn't be deleted using the the "extremely poor quality" flag mechanism that already exists, assuming that it had no redeeming qualities? \$\endgroup\$ – Scott Seidman Aug 28 '18 at 12:21
  • \$\begingroup\$ ... and you said "I think" ;-) \$\endgroup\$ – Scott Seidman Aug 28 '18 at 12:21
  • \$\begingroup\$ No, it needs an explanation rather than just poor quality. And of course I said I think. It was a comment, not an answer, and this is a discussion, so obviously it is an opinion based thing. That should have been obvious, thus no need for that pointless comment you added \$\endgroup\$ – MCG Aug 28 '18 at 12:34
  • \$\begingroup\$ @ScottSeidman look at the answer posted on a question today. This is not a direct 'I think' answer, but it is, however a guess, which is the main point of this post, it wasn't all 'I think' answers, it was against guesses with no evidence to back it up. This falls under the guess category, but doesn't say 'I think' anywhere. Hence my downvote, flag, and comment. electronics.stackexchange.com/questions/393059/… \$\endgroup\$ – MCG Aug 28 '18 at 13:52
  • \$\begingroup\$ @MCG -- it looks like it's been 100% appropriately handled. I see no need for any change. \$\endgroup\$ – Scott Seidman Aug 28 '18 at 14:04
  • \$\begingroup\$ @ScottSeidman Those were the sort of answers I was trying to speak against. Guesses with no evidence \$\endgroup\$ – MCG Aug 28 '18 at 14:17

You must log in to answer this question.

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