There are lots of governments (including mine) that deliberately fabricate data, statistics, investigation reports to advance their agenda. How should we examine these claims?

3 Answers 3


If you have proof that a report is fabricated, with good references and all the good thingies we at skeptics like, feel free to challenge the claim that is presented on the report.

If you don't have proof - and by proof I mean more than "my government usually lies about stuff" - then you can't really challenge it. It would be your world against theirs, and normally a government has more leverage with skeptics than a single person.

This of course doesn't apply to philosophical or religious claims made by any government. If, let's say, a random christian nation says that 78% of teenagers would end up in hell because of some sort of gaming habit you can't really take the paper seriously for skeptics matters.

Otherwise, for police reports, economical data, census, etc. we can give a bit more of trust - not blind trust, mind you, but they are definitely worth considering and more powerful than some random blogger out there or some holocaust-denial nutcase screaming at the top of a soapbox.

As anything on skeptics - give us proof and/or references. If you have proof that a report is wrong, it will be welcome to the site. Otherwise, the community won't really take you seriously on your claims.


In questions, feel free to ask about those claims. Governments are generally not unique sources except some exceptional things.

I don't see a problem using government reports in answers either, unless there's reason to be skeptical about them (use voting to agree and disagree with the appropriateness).

  • Sorry, -1. Claimes by a biased source that has BOTH the motive, the opportunity, AND the reputation for lying, is a really really poor source - and yet, because it's "official government", it's automatically extra "solid" to many readers because of Authority bias (probably a better bias can be mached, since Authority implies unrelated expertise?).
    – user5341
    Jan 10, 2017 at 4:01
  • @user5341 I believe you misunderstood me, answers based on reports by governments need not be about claims by governments
    – Sklivvz
    Jan 10, 2017 at 9:32
  • I don't care about the claim's topic. My problem is with using government claim as the answer's source (without openly and clearly stating it's likely to be fake, or even mis-used)
    – user5341
    Jan 10, 2017 at 12:06
  • @user5341 We are still talking past each other. There are innumerable government papers which are completely reliable.
    – Sklivvz
    Jan 10, 2017 at 12:10
  • From Iraq or Cuba government? Doubtfully.
    – user5341
    Jan 10, 2017 at 12:19
  • 1
    @user5341 and who decides which government are OK and which other are not OK? The users, via voting, just like any other resource.
    – Sklivvz
    Jan 10, 2017 at 12:48
  • The users, from observed practice, tend to severely cognitively overcount (is the the opposite of discount) the tendency of governments to cook books and misrepresent data - especially in countries without adversarial press; and treat it as highly as randomised double blind control trial results. Especially when it fits their worldview (heck, I do the same - it's normal cognitive bias)
    – user5341
    Jan 10, 2017 at 14:35
  • I found government sources to be highly unreliable when it comes to libertarian issues. Prostitution, drugs, porn, things that government criminalize but actually victimless. Governments would insist and exaggerate how "dangerous" things are.
    – user4951
    Sep 26, 2018 at 5:05
  • @J.Chang I am not sure I agree. You are referring to the US government and internal US political positions, neither of which have any major relevance to my answer. Usually government publications are well sourced, so that's at least a step above most other publications.
    – Sklivvz
    Sep 26, 2018 at 7:16
  • drugabuse.gov/publications/drugfacts/marijuana I would say this is accurate enough. Pretty objective. Sometimes the government says "can" while in reality "can" could mean 1 in 1 million or 1 in 10. So a bit misleading. What do you think?
    – user4951
    Sep 26, 2018 at 7:56
  • @J.Chang I think you don't agree with that page, perhaps, and this biases your view, but it's a well referenced page. I think it makes a valuable secondary source. In the range of all the possible levels of being accurate and reliable, this ranks way better than most pages on the internet. While you might see it as misleading, at least you have references to read. So you can find out that a "can" might mean "1-in-a-million chance" by reading the sources.
    – Sklivvz
    Sep 26, 2018 at 8:41
  • Actually I agree with you. I think that page is, surprisingly to me, reasonably objective. Basically I think ganja is far less dangerous than alcohol and cigarettes and not even government sources are willing to dispute that. I wonder if US government are also honest in anything else? Does it?
    – user4951
    Sep 27, 2018 at 8:57
  • Saying that smoking weed can cause permanent damage in your brain is like saying walking can cause permanent damage to your toes. In a sense, anything is possible. That's technically correct there. But we probably mean serious damage with significant probability that worth considering.
    – user4951
    Sep 27, 2018 at 8:59

I don't see how this is any different than any other topic on this site. You see a stat, are skeptical, ask a question. But you may have to give a decent reason why you're skeptical about official reports. I don't think "my government usually lies" counts. "Many have little faith in my government's reports, so third party verification would be nice" probably does count.

You must log in to answer this question.

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