Possible Duplicate:
Discussion on FAQ Question: What are the attributes of a good answer?

What is a good answer? And what is a "bad" answer, within the context of this forum? What are we looking for in an answer: what makes for a "good" answer?

I ask because, sometimes there are answers (especially from new users) which don't follow the conventions of the forum: and I'd like to be able to reply to these answers with a comment that includes a link to a FAQ, which describes what kind of answer is expected or requested.

  • Apparently the answer I cited in my question has been deleted. Oh well.
    – ChrisW
    Jun 17, 2011 at 5:41
  • I didn't see that answer when it was submitted, just out of curiosity you mind giving me the gist of it? Jun 17, 2011 at 14:47
  • 1
    I was just about to ask a similar question, as I seriously keep rubbing my eyes at the speculative, anecdotal answers that keep coming up... even from folks with ~500 reputation! There has to be something unclear about expectations going on.
    – Hendy
    Jun 17, 2011 at 21:27

3 Answers 3


FWIW, the answers to How much energy does displaying a webpage with a black background actually save? probably have the largest dissonance with regards to claimed values and community response. The top two answers (currently at +82/-2 and +59/-3) both contain no references and original research. The comments on both answers show a big split on opinion.

I don't have much to say in direct response, but this is a good case study of where the borders are.

  • 1
    I would add that the "original research" there was toward the objective side in the case of using a Kill-A-Watt. In other words, the methodology is published, monitor models published, and results recorded. Anyone is free to replicate. I don't see this as that different than Mythbusters doing the same thing and us thinking their answer was reliable. And this is insanely different than "my monitor feels warmer on blackle than google."
    – Hendy
    Jun 17, 2011 at 21:25
  • @Hendy: Agreed. To be clear, I am simply presenting this as an interesting case.
    – MrHen
    Jun 17, 2011 at 21:49
  • 1
    This question is on me to-discuss list. I simply have not had the time to deal with it yet.
    – Borror0
    Jun 17, 2011 at 21:54
  • makes sense.
    – Hendy
    Jun 17, 2011 at 22:04
  • I usually neither up nor downvote answers that are like that. We've gotten some anecdotal answers that are really interesting to read. In those very rare cases I think we would be worse off deleting really interesting anecdotes like skeptics.stackexchange.com/questions/2024/… even though they are poor answers. So not voting (or flagging) on them has been my own solution so far.
    – Kit Sunde
    Jun 18, 2011 at 1:43

There are many components to a good answer on Skeptics.

1. Void of speculation.

The primary purpose of Skeptics is to verify or falsify claims. As such, speculation is to be avoided at all costs as it unhelpful at verifying or falsifying anything. If an answer an be summed to "I don't know for sure, but here is what I think the answer is," then it is a bad answer.

2. References any significant claim.

At the present time, we have a strict "no original research" policy.

Simply put, hearsay and anecdotes are not enough. All significant claims have to be backed up by a reference that has credibility in the domain (i.e. no encyclopedia, no source which may be biased, etc.) and preferably peer-reviewed literature.

Obviously, not all claims have to be backed up. If the question can be answered with high school level science, no references are needed. The asker can always browse Wikipedia if is still skeptical.

3. Explains the methodology.

Citing studies is not enough. While a study may prove that X has a correlation with Y, the methodology is important - especially when it comes to social sciences where different definition of terms like "happiness" or generosity" would change the answer drastically.

4. Written in plain words.

Most Stack Exchange sites are intended for expert-to-expert exchange. Skeptics is different in that most users will be laymen in the topic discussed, be it biology, physics or psychology. For this reason, an answer should always be written in a plain language void of overly technical terms so that it might be understood my the average user.

5. Written in a polite and neutral tone.

As Robert Cartiano once said, "'[a] bunch of dicks' equals a failed site. It's really as simple as that."

We expect users to be respectful of others they disagree with, no matter how frustrated they may be. Rants about young earth creationists, "climate change denialists," etc. will not be tolerated here. If we want to introduce people to the evidence, we have to be nice. Otherwise, we'll drive them away before we even got a chance to educate them.

  • Let me know if I forgot anything. I'm sure I have.
    – Borror0
    Jun 17, 2011 at 22:15
  • It should be on-topic, i.e. answering the question and nothing else: see Shog9's comment on this answer.
    – ChrisW
    Jun 18, 2011 at 1:18
  • You could, optionally, make this question CW and then format each of those topics as a separate answer: so that meta-users can comment on, question, edit, and add to specific topic; so that novice answerers can be pointed to the specific topic which their answer shows they're unaware of.
    – ChrisW
    Jun 18, 2011 at 1:24
  • @ChrisW: I love the idea. They did something similar on Science Fiction: meta.scifi.stackexchange.com/questions/350/… Since this answer has a two other answers, I have created another question though: meta.skeptics.stackexchange.com/questions/868
    – Borror0
    Jun 18, 2011 at 1:42

on stackoverflow i saw some answers accepted as solution, but not accepted later given answers with a much higher vote up.

The problem on skeptics.SE is, if a answer was accepted, nobody will really try to make a better one. The questioner and his level of understanding practically defines what a good answer is. I tried here on some question to give a more critical answer, to say what question might be better, more specified, what can be said at all. But the simple direct answer with much links wins nearly always.

If this can be overcome by more users, i dont know, have my doubts after reading some of meta.physics.SE discussions which had similar problems. I think more competition here will not really drastically raise the quality. We should concentrate on getting more good questions imo. Seems much easier and better to control. A good question mostly forces precise answers. The question on torturing showed that political correct can also mean good, whether the answer quote quite objective people/studies or not. Similar to your criterions on good questions, you will imo not find really common standards, the best is a bottom line of allowed questions.

But mods seem here to go the way "lets grow and popularize skeptics.SE first, than we can raise the quality of Q&A". If you look on meta.physics discussion drawing parallels to mathoverflow, math.SE the term broken window come up which yields quite the opposite. If you dont can establish a quality of Q&A (mainly through good moderation, make it an interesting place for AT LEAST some experts) from the very beginning, you will not achieve a "good" Q&A level at any time. Then this experiment here ends just like another YahooAnswers.

In this context the rating system here has not the task to rate answers good, but to reward experts and diligent users. Viceversa on stackoverflow the dominant function of the rating system is to rate answers good, as there were from the beginning many experts on this site as on mathoverflow. Then actually the rating system works. Here it has in principle many downsides as explained.

Some imo very good reads/discussion/esp. look at the comments

https://physics.meta.stackexchange.com/questions/76/level-of-questions https://physics.meta.stackexchange.com/questions/94/level-of-questions-what-do-we-do-now

  • 5
    "mods seem here to go the way 'lets grow and popularize skeptics.SE first, than we can raise the quality of Q&A'" That isn't accurate.
    – Borror0
    Jun 17, 2011 at 21:20

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