There are some studies performed by political organizations or NGOs with sole role of validating their point of view or getting their hands on taxpayers' money.
These studies are heavily biased and one sided.

Should these be accepted as valid sources?

  • 2
    There has already been an attempt to qualify what makes a good reference. Do you feel this concern is not covered more generally by that? – Oddthinking Jun 30 '11 at 12:23
  • Well, linked meta-answer describes ideal situation. I'm asking, because I have seen quite a few answers reference biased sources. – vartec Jun 30 '11 at 12:47
  • a) Which political organizations do you talk about? b) which NGOs do you have in mind? c) can you prove your claim about the sole role? d) Doesn't getting taxpayers money already contradict the previous claim of the sole role? e) Heavily biased are your anonymous accusations. Please substantiate your claims, and assure us, that you neither work for a political party or institute, nor for an economical interested concern, a private research lab or somehow interested publisher, or are a member of a NGO, church or ideological biased group. Thanks. ;) – user unknown Jun 30 '11 at 14:43

Fabian gives a good answer from the answerer's perspective.

From the reader's perspective:

No, you shouldn't accept sources from others that are demonstrably wrong in their answers, due to biased analysis, or that you perceive might be unfairly biased due to an undisclosed conflict of interest.

Your remedies include:

  • Downvoting the answer.
  • Commenting on why you think it should be downvoted until fixed - be specific and be nice.
  • Upvoting other comments on the problem, so the issue doesn't get hidden and other voters can see it.
  • Downvoting the answer.
  • Editing the answer to disclose the conflict of interest, so others can judge whether it is a factor.
  • Editing the answer to add a better reference (assuming the same conclusion)
  • Downvoting the answer.
  • Writing a competing answer
  • Also, consider downvoting the answer.

From the moderator's perspective:

If by not "accepting" you mean the moderators should simply delete the answer like we would if it contained profanity, then no, I don't think that is necessary as a site rule. Better if the community can take care of it.


Our rules on requiring references are just the minimum necessary for an acceptable answer, just because you include some references does not mean you have composed a good answer. The references can be outright wrong, misleading or you might just have interpreted them wrong. But having a reference is far better than not having it as other users can examine it and judge it for themselves.

Our policy on requiring references is just a very basic measure to filter out the almost certainly useless answers. Judging the reliability of a reference is the obligation of the author and in part of the other users reading that answer.

If there is an obvious conflict of interest in a study you cite, you should point that out so that the readers can take this information into account. You should critically examine the methodology of the studies you cite if you have the necessary expertise to do that. You also should not stop researching just because you found one reference, having multiple references that arrived at the same conclusion by different means can greatly improve the reliability of your answer. If you have reason to doubt your only reference, by all means try to find a better reference.

I consider it acceptable to use studies from potentially biased sources, but the conflict of interest should be mentioned in the answer and the reliability of the source should be examined.


Three answers:

  • IMO yes they 'can' (not necessarily 'should') be considered 'valid' or 'accepted': BUT if there are other, conflicting/opposing answers, then beware of mentioning only one of these answers, only one side of the argument. I'm not saying that if you give the mainstream answer then you should give every other crackpot theory too, but conversely I am saying that if you give a (IMO) crackpot theory then you should also at least mention and ideally address the more mainstream answer[s]. I try to summarise this opinion in this meta-answer ('It should summarize the whole truth'), which was inspired by the down-voted answers to the Is Soy bad for you? and Does smoking under 3 cigarettes a day prevent addiction? questions. It's not the fact that a source is biased that would worry me: what would worry me more is if their message were counter-factual or contradicted.

  • For an alternative point of view (not necessarily mine), see also Does 'authoritative' mean something? which may be a question about the extent to which the source of the information (rather than the information itself) may be important.

  • See also this meta-answer ('It must explain the methodology of the studies it cites') which may help to mitigate some 'bias' which may be built-in to even a truthful answer


Pharma companies make studies with the sole purpose of getting drugs approved by the FDA. The point of the study is to make money. Does that mean that we should forbid people from linking to those studies?

No. The existence of those studies is still important for people who want to form an opinion about the drug. It makes sense to references them.

The same goes for NGO that have ideological interests instead of commercial interests.

You must log in to answer this question.

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