Here are some standards of proof from Wikipedia "Legal Burden of Proof", listed from easiest to most stringent:

Some credible evidence...

The "Some Credible Evidence" standard does not require the factfinder to weigh conflicting evidence, merely requiring the investigator to present the bare minimum.

Preponderance of the evidence...

Preponderance of the evidence, also known as balance of probabilities is the standard required in most civil cases. The standard is met if the proposition is more likely to be true than not true.

Clear and convincing evidence...

To prove something by "Clear and Convincing Evidence", the party with the burden of proof must convince that it is substantially more likely than not that the thing is in fact true. This is a lesser requirement than "Proof Beyond a Reasonable Doubt", which requires that the trier of fact be close to certain of the truth of the matter asserted, but a stricter requirement than proof by "Preponderance of the Evidence," which merely requires that the matter asserted pass the 50% threshold of being more likely true than not.

Beyond a reasonable doubt.... This high standard is used in criminal court because someone's freedom is at stake....

It has been described as, in negative terms, as a proof having been met if there is no plausible reason to believe otherwise.If there is a real doubt, based upon reason and common sense after careful and impartial consideration of all the evidence, or lack of evidence, in a case, then the level of proof has not been met. Proof beyond a reasonable doubt, therefore, is proof of such a convincing character that you would be willing to rely and act upon it without hesitation in the most important of your own affairs. However, it does not mean an absolute certainty.

Beyond any imaginable doubt This one isn't in the wikipedia article, but describes a game some people play here.

Wikipedia does discuss Beyond the shadow of a doubt

Beyond the shadow of a doubt is the most strict standard of proof. It requires that there be no doubt as to the issue. Widely considered an impossible standard, a situation stemming from the nature of knowledge itself, it is valuable to mention only as a comment on the fact that evidence in a court never need (nor can) reach this level.

While Skeptics is not court -- no one gets paid any money, and no one goes to prison -- there is also no point in reinventing the wheel if one or more of these standards of proof can apply in understanding our duties to upvote only the best answers and downvote the poor answers.

It seems we have some participants who use "enjoyed the post" or "Some credible argument" (not Some Credible evidence..., since evidence means a citation) as their standard. Preponderance, without evidence, is also pretty close to "a credible argument". I even confess to have upvoted some of these. Others are using Beyond the shadow of a doubt or Beyond any imaginable doubt (e.g. have we REALLY been to the moon, what if there is no moon? what if the gubermint lies? the earth could be created 4000 years old).

Somehow, we need to convince users to adjust their standards towards the upper middle (e.g. clear and convincing evidence) of this scale or define some other standard of proof not shown here.

Is Clear and convincing evidence the best standard to adopt on Skeptics, or is some other standard more appropriate? (either picked from the above list or invented here)

  • Quoting from faq:" What is Skepticism? It's strongly related to science and the null hypothesis; that is, everything is false until proven true through strong, verifiable evidence. " – Theta30 May 6 '11 at 7:29

This is not the right approach.

Different disciplines will have different standards. An historical "proof" cannot be as convincing as a chemistry "proof". Hard science experiments must be reproducible, other sciences can only examine existing evidence. Some statistical sciences can only examine correlations. The standards of proof differ from field to field, and from case to case.

I think that the important bit is to disclose the level of confidence we can attribute to a claim.

We are not interested in passing judgement, but only to present and summarize the available evidence.


What I'd like is "clear and convincing evidence", but in some cases that may be too strict for an individual answer.

We don't actually need a single, non-collaborative, answer that provides the evidence and shows the weakness of opposing evidence. We don't want to encourage people to take somebody else's pretty good answer and copy or paraphrase it in its entirety just to add some more supporting evidence. Therefore, we need to accept answers that are part of clear and convincing evidence.

Answers should therefore be acceptable if they help significantly in forming "clear and convincing evidence". This means that the minimal standard for acceptability is "some credible evidence", but not all of that would be acceptable. Such answers are useful only when they supplement other answers; if the "some credible evidence" comes first, it should be clear that it can usefully supplement other answers.

There are subjects where there just isn't enough evidence extant to reach "clear and convincing". If we require such answers, then such questions must go unanswered, and I don't think that's ideal either. Nor do we necessarily know that there is no clear and convincing evidence, so we can't show that.

So, my proposal is a minimum of "some credible evidence", preferably "preponderance of the evidence", and to be acceptable such answers must have the potential of being part of a larger "clear and convincing" answer.


If I ask: "Is A true or false?" than the goal of an answer isn't to show that A is true beyond a reasonable doubt. The goal is to provide an argument that A is true or false with might use facts X, Y and Z. All the facts should have substance. If you could include a fact in a academic publication it has substance.

If an issue is still contested and there a good argument to be made for either case it would be ideal to have answers that summarize all relevant arguments that can be made on both sides of the issue. It's okay to have an upvoted answer that says A is true and an upvoted answer that says A is false provided both are making a argument based on facts.

You must log in to answer this question.

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