Today we had a question that unleashed a torrent of opinion-based answers. If evolution is not about increased complexity, why does so much complexity evolve?

While a few of them were factually correct, only one answer, coming from an experienced user of the site was referenced.

There have been two POB votes on it for most of the day, while other questions that have not received answers have been closed within that time. To me this means that even though the answers clearly demonstrated that the question yielded not only primarily, but almost exclusively, opinion-based answers, there was no consensus that the question should be closed.

So is the Primarily Opinion-Based vote to close only there to censor questions we do not like or are we willing to enforce it when the question clearly demonstrates that the answers generated are opinion-based, even if the question itself may warrant some thoughtful response?

At present there are ten answers and only one has any references what-so-ever. The top rated answer has 29 upvotes and one downvotes and ignored the request to improve the answer with references.

I would like to understand how the site wants to proceed?

  • $\begingroup$ Some of the POBs... "Humans are an apex predator... The only reason we have iPhones is because intelligence is the ultimate hunting weapon." biology.stackexchange.com/a/42086/16651, "It might help to not think about evolution as a process at all - it tends to imply some sort of planning or goals or something like that." biology.stackexchange.com/a/42064/16651, "I'm going to chime in here" biology.stackexchange.com/a/42071/16651, "I don't know the details of the simulations you are referring to but I'll give my answer anyway." biology.stackexchange.com/a/42055/16651 $\endgroup$
    – AMR
    Jan 7, 2016 at 3:17
  • $\begingroup$ "Computers are dumb. They can only do what you tell them to. The universe is not dumb, it has a plethora of mechanics and interactions that are far too many to simulate with even all the computing power we have combined." biology.stackexchange.com/a/42057/16651, "Evolution is about improvement, but not about improvement of the species in general, but about improvement of the species in the current environment to get children and their children to get children and so on." biology.stackexchange.com/a/42068/16651 $\endgroup$
    – AMR
    Jan 7, 2016 at 3:19
  • 1
    $\begingroup$ Judging by the view count, that question made it into the Hot Network Questions list. All bets are off at that point anyway, this inevitably leads to distorted voting and often to mediocre answers in exchange for a boost in traffic. This can be annoying, but it's also not the end of the world and I'd be careful not to use these exceptional cases alone to justify global site policy. $\endgroup$ Jan 7, 2016 at 19:47
  • $\begingroup$ @MadScientist The outliers provide the clearest examples. The SE model doesn't really allow for community valuation of answers in the same way that it does questions. There is no close voting on answers, and even completely opinion-based answers wouldn't be grounds for deletion under the guidelines. If we go by what SO says, then it is the question that we go after to protect the quality of the answers. blog.stackoverflow.com/2010/09/good-subjective-bad-subjective $\endgroup$
    – AMR
    Jan 7, 2016 at 19:55
  • $\begingroup$ I don't think that questions should be evaluated based on the quality of the answers they are getting. A question can be "good" and not opinion-based, but might still get poor opinion-based answers that lack references. Opinion-based answers might be an indication that the question has problems, but could just as well signal lazy answerers. $\endgroup$ Jan 8, 2016 at 21:21
  • $\begingroup$ @fileunderwater I think it gets down to this [SO blog post](blog.stackoverflow.com/2010/09/good-subjective-bad-subjective), where we know the questions that will be problematic and those that will either be answered well or will have unreferenced answers only because of people's laziness. $\endgroup$
    – AMR
    Jan 9, 2016 at 1:48

2 Answers 2


Here's my current thoughts on this. I think it's wrong to be closing questions that have a reasonable chance of getting a quality answer. The responsibility of providing an answer which is factual, objective, and supported by evidence falls primarily to the user posting an answer, and secondarily to the wider community. Of course, the wording of some questions can be improved to make them less inviting to opionated answers.

I think we should be using the POB close vote when there is little/no chance of getting an answer that is objective/factual, not because there is a high chance of opinionated answers.

There are two ways one can interpret the reasoning for closure:

"Many good questions generate some degree of opinion based on expert experience, but answers to this question will tend to be almost entirely based on opinions, rather than facts, references, or specific expertise."

I would say can be interpretted as either

  • Most answers will be just opinion pieces

  • All answers will be largely opinion with little/no factual evidence

I think the latter is what we should be using this closure vote for, otherwise we are potentially preventing perfectly good questions from being answered because they are the sort of question that attract opinionated users. E.g. "Is evolution real?" Can very much be answered in a factual objective way but, as we've seen before, will attract a large number of opionated posts. It is for the benefit of the scientific community to provide a valid answer to such questions.

Questions that can only be answered with opinion are not suitable for the Q & A format, and have lesser value. For example, "Is Excel or R better for analysing my data from my experiment?" or "Which is the the best quantitative genetics book?" are only going to give opinionated answers (PS. the answers are R and Falconer and Mackay or R and Lynch and Walsh)

What this means though is we need to be commenting on, downvoting, flagging and deleting low quality opinionated answers. Happily though, this will really help maintain the high quality of answer posts on biology SE!

  • $\begingroup$ There is a problem with the plan as it comes to moderating answers. Upvoted answers can only be deleted by Moderators, and there is no mechanism in place for the community to cast close votes to answers. Even at 20K, the answer needs to be downvoted, and it has to clearly exhibit that is not attempting to answer the question. Things such as "HAHA Look I can poop!" biology.stackexchange.com/help/privileges/trusted-user $\endgroup$
    – AMR
    Jan 7, 2016 at 19:43
  • $\begingroup$ In my answer to Matt Thrower's meta post, I basically showed where I would have tried to add references. The answer as it stood could have benefitted from citations to bring it in line, but I don't think that it would have fallen under the guidelines to delete, actually most of the answers, including the accepted answer, isn't clear that they should be deleted. Does ignoring requests for citations produce grounds for deletion? I think that would be a much more difficult call than putting a hold on questions that will generate a majority of responses that will not be well researched. $\endgroup$
    – AMR
    Jan 7, 2016 at 19:49
  • $\begingroup$ Perhaps we can lobby stack exchange for new functionality - I think we are a site that is prone to opinion answers more than most sites, having the ability to control such answer quality in a similar way to the question closure system might be useful $\endgroup$
    – rg255
    Jan 7, 2016 at 20:15
  • $\begingroup$ I only think that you see a few specific types of questions that get opinion answers. I think that most regulars provide really well referenced questions. Off the top of my head Christiaan and Anongoodnurse recent answers have been textbook. $\endgroup$
    – AMR
    Jan 7, 2016 at 20:43
  • 1
    $\begingroup$ Can we maybe get at this more like how Wikipedia does it? Go in and edit answers with [citation needed] brackets every place in an answer where opinions require reference or evidence to back up, even if maybe it might be common knowledge to people in the field. It may not change the behavior or get a response, but it will at least show people reading the answers that they are not up to community standard? $\endgroup$
    – AMR
    Jan 7, 2016 at 21:01
  • 1
    $\begingroup$ I have made an experimental edit to biology.stackexchange.com/questions/42050/… - I think whatever we write we should be prepared to back it up with citations if requested. E.g. I can write that the four mechanisms of evolution are selection drift mutation and migration, and I wouldn't generally think to put a reference, but if someone asked me to do so I would. If it's such common knowledge that one doesn't think a citation is necessary it should also be easy to provide one! $\endgroup$
    – rg255
    Jan 8, 2016 at 10:59
  • $\begingroup$ @AMR I definitely see the purpose of asking for citations, but the risk of wikipedia-stype [citation needed] (as rg255 also added to the linked A) is that it they can easily lead to edit wars between users, that can spill over to other Qs and As ("revenge edits" asking for citations, just to annoy or done in spite). At SE, people generally edit other answers to improve them (i.e. edits are done in good faith), and people edit answers they think are good but that can be improved. In contrast, I imagine that [citation needed]-requests will be added to answers that people dislike (~"bad faith"). $\endgroup$ Jan 12, 2016 at 9:34
  • $\begingroup$ I don't think a citation needed approach will work. As Fileunderwater says, edits should improve the post, and it could lead to an easy route for petty actions (revenge edits etc.) - but it's clear we need to find a way of dealing with low quality answers, including unreferenced material where users ignore or refuse requests to cite - is that grounds to flag and delete answers?? The citation needed approach does at least make it very clear what citations are wanted $\endgroup$
    – rg255
    Jan 12, 2016 at 9:41
  • 2
    $\begingroup$ The moderators have the ability to add this banner to the posts. "Some of the information contained in this post requires additional references. Please edit to add citations to reliable sources that support the assertions made here. Unsourced material may be disputed or deleted." maybe it would be worthwhile to put in a request that functionality be added to make that a community vote for answers, as is the case with close votes. It would reduce Mod flags, and would provide a way to highlight that the answer should be read critically as it may not be backed up by facts. cc @fileunderwater $\endgroup$
    – AMR
    Jan 14, 2016 at 20:32
  • $\begingroup$ @AMR Yes, that banner could probably be used a lot more. To let user vote for such banners on answers could definitely be useful, but probably unlikely to be implemented across all SE sites (which I suspect would be needed). $\endgroup$ Mar 30, 2016 at 8:17

So is the Primarily Opinion-Based vote to close only there to censor questions we do not like or are we willing to enforce it when the question clearly demonstrates that the answers generated are opinion-based, even if the question itself may warrant some thoughtful response?

Closing a question as POB should be based only on the fact that there is no scientific answer to a question. To close a question because the answers are POB is like throwing the baby out with the bath water.

I'm not sure this answers your question well, but my belief is that POB is used too much. Science is a very broad subject, and many times "POB" is used instead of "we don't know the answer to this question." The problem with that is often, someone knows. They just haven't read the question or answered it.

When I first came here, I saw a lot of medical questions closed as POB. That's one of the reasons I almost always include a study or some decent reference(s) to support my answer. This is, after all, a science site.*

If you want to shut down POB answers, hit them with post notices, comments asking for sources, and, when appropriate, down votes. But don't close the question for the wrong reason.

*I admit I've been a bit more lax of late. But I have 189 answers here, and I bet 160 or more of them have references.

  • $\begingroup$ I started another post, that you saw and commented on already that is basically a feature request to allow the community a way to address answers and not just the questions. That is probably the better approach, The question, as you said may have a scientific answer, but it will generate a lot of bad answers, which may not necessarily be the fault of the question. There is the mechanism to request a question be reopened, but it is likely better to leave open and moderate the answers instead of the question itself, unless it is flame bait or will clearly only have POB answers. $\endgroup$
    – AMR
    Jan 15, 2016 at 8:41
  • $\begingroup$ I firmly believe that's the way to go. It's also a burden to the folks who answer, but if they know their stuff - and that's the folks you want answering questions - then a reference or two isn't too difficult to find. $\endgroup$ Jan 15, 2016 at 8:46
  • 1
    $\begingroup$ I usually check sources to make sure I have remembered something correctly, and if I really don't feel like putting in the effort, I put in a comment to point them in the right direction. $\endgroup$
    – AMR
    Jan 15, 2016 at 8:49
  • $\begingroup$ @AMR - That's a reference in my book: something that backs up your answer or provides a place to read more. $\endgroup$ Jan 15, 2016 at 8:52
  • $\begingroup$ I think this is a good post on the matter. Am I right to think then that this is generally agreeing with what I've also written? i.e. if it seems there's actually a legitimate answer which can be given then the question should remain open, and the responsibility falls to ensuring the answers are up to scratch? Here's some important excerpts from my answer: Interpretting POB as All answers will be largely opinion with little/no factual evidence and regarding opinionated answers we need to be commenting on, downvoting, flagging and deleting low quality opinionated answers. $\endgroup$
    – rg255
    Jan 15, 2016 at 11:31

You must log in to answer this question.

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