I am writing a PhD thesis in robotics and I was wondering if it sounds good to start a phrase with: 'According to Wikipedia ...'

Do you think I can do that or not?

  • 13
    You can... but doesn't mean you should. There is no force on Earth that will stop you from writing those words.
    – Bluebird
    Oct 18, 2017 at 10:24
  • 2
    Don’t quote Wikipedia.
    – Bluebird
    Oct 18, 2017 at 12:53
  • 3
    obligatory xkcd
    – Pac0
    Oct 18, 2017 at 15:40
  • 4
    Wikipedia articles usually refer to published papers to sustain claims. You should cite those instead.
    – a06e
    Oct 18, 2017 at 15:47
  • 7
    That sounds like a phrase from a half-baked high school essay, not a PhD thesis. If you want to be a scholar, you should read and cite the primary literature. Oct 18, 2017 at 16:09

8 Answers 8


"According to some unknown guy on the Internet, I can say that according to Wikipedia..."

No, you should avoid that. Wikipedia is open, so it can be changed by (almost) everyone and you have no guarantee that the given information there is correct. While many articles are very informative, it might be better to take the actual source (given on the Wikipedia page) and make sure that this source is credible (e.g. published in a good, peer reviewed journal).

  • 17
    Apart from the obvious truth in this answer, the only situation I could imagine citing Wikipedia like that would be to clarify some common misconceptions stated in the wiki-article. "According to wiki, XYZ is a typical source of ABC, representative for a common misunderstanding in non-scientific literature about ABC-XYZ interaction, originating from an faulty interpretation of early DEF-experiments. Here, I want to address..."
    – BPND
    Oct 18, 2017 at 10:16
  • 10
    Another case where one could actually cite Wikipedia is if you want to cite data explicitly from Wikipedia (e.g. some information on the number of available articles, on the number of readers, etc.).
    – Dirk
    Oct 18, 2017 at 10:25
  • 9
    Actually, Wikipedia is a good, peer reviewed site. And the information is probably more accurate than in the average journal article. Oct 18, 2017 at 13:14
  • 3
    @JiK I can't speak for Meier, but I personally learned that from the meticulously detailed, well sourced, and impressively unbiased Wikipedia article on the topic: en.wikipedia.org/wiki/Reliability_of_Wikipedia Oct 18, 2017 at 15:19
  • 4
    @JiK I know that Wikipeda has a bad standing in academia, and I know that citing XY journal is usually considered to be "better" than citing Wikipedia. Nevertheless I know that many journal papers are only partially read by the referees - if an error is found later on, it is rare that any kind of erratum is published. Wikipedia, on the other hand, is read and corrected by many people and the chance that is right is therefore higher. Oct 18, 2017 at 15:21

I think that Dirk’s answer is highly misleading and represents a popular, but false view of the website. Investigations have shown that Wikipedia contains as many or fewer statements of incorrect fact on scientific and technical matters as more traditional encyclopedias like Encyclopedia Britannica (though "mistakes by omission" are more common on Wikipedia). Just because Wikipedia can be edited by anyone doesn’t mean it’s unreliable. Additionally, just because anyone can add content doesn’t mean that there are no controls on content, nor does it mean that there aren’t people or systems for double checking edits.

However, it is correct that one shouldn’t cite Wikipedia because Wikipedia (like most encyclopedias) is a tertiary source and has a strict “no original research” policy. Encyclopedias are summaries of information, but don't contain original research by design and are not the original source of the content. You should always cite the original source of the content, not a reproduction of it. If you wish to credit an encyclopedia with drawing your attention to the material, you can append "accessed via Wikipedia" to a citation, which I've seen done in some published scholarly books and journal articles.

  • 1
    I doubt Dirk's argument is that Wikipedia is unreliable because it can be edited by anyone. Rather, because it can be edited by anyone, referencing Wikipedia is equivalent to 'according to some guy on the internet' and therefore serves little purpose as a citation, regardless of whether the information is correct. Oct 18, 2017 at 14:54
  • 1
    @MarcksThomas I disagree about your interpretation, but let's say it's correct for a second. Citing "some guy on the internet" who writes a well articulated, well analyzed, and well sourced blog article is completely reasonable. Quality work is quality work, even if it doesn't have "Dr. Mathematician, PhD" attached to it. I have personally submitted research papers to the US government that cited multiple blog article. Those blog articles were excellently written and had the best analysis of the relevant data. Oct 18, 2017 at 15:16
  • 2
    I am not sure why a comparison to Encyclopedia Brittanica is relevant. I would also find it strange to see a reference to that. Oct 18, 2017 at 15:49
  • 3
    @TobiasKildetoft That's the point of my answer. I'm saying that encyclopedias are inappropriate sources for citations because you should always cite primary sources. Oct 18, 2017 at 16:01
  • 2
    For the record, I'd find "According to Encylopedia Britannica..." also strange to see in a PhD thesis.
    – Fomite
    Oct 19, 2017 at 19:09

A good Wikipedia article must be supported with credible sources and references (see here and here). Wikipedia itself has a good read on citing Wikipedia.

In case of scientific articles, the references are usually either published books and/or articles and papers from scientific journals and conferences. In either case, you should be able to follow them and reference the Wikipedia article (or the section of your choice) back to any of these original sources, and then cite that original source in your thesis. That makes for a valid and accepted citation instead of something that someone might have randomly written on the day you checked the article.

  • 1
    (+1) But what would one do when you learn something very basic on Wikipedia (which is probably outside one's field) that does not require citation? For example, "According to Wikipedia, the chain rule is ...".
    – Cliff AB
    Oct 18, 2017 at 14:21
  • 2
    @CliffAB find a credible source.
    – msouth
    Oct 18, 2017 at 14:36
  • @CliffAB In most, if not all, fields, even basic subjects are well-defined in a credible textbook. For example, the "chain rule" must have already been explained in at least one reference textbook of the related academic or engineering field (and should be referenced properly in the Wikipedia article). Since original content is not suggested to be used on Wikipedia, even basic topics shouldn't be written based on "common" or "personal" knowledge.
    – ali14
    Oct 18, 2017 at 14:36
  • @msouth: My point about the chain rule is that no mathematician would ever need to cite the chain rule in a paper. However, if you're publishing in a field that traditionally has no exposure to math (lit?), and for some reason you find the chain rule very helpful, it might be a good place to put a citation of the chain rule...even though you learned from a Wikipedia page, and pointing readers to the citations given by the Wikipedia page would not be as helpful as the Wikipedia page itself.
    – Cliff AB
    Nov 1, 2017 at 16:43
  • @CliffAB This is about a PhD thesis--I would cite a book. You're trying to tell someone where the data is that backs up your claim. You don't want to reference something that could change thousands of times between when you wrote it and the last time anyone ever looks at your references. I think a website is iffy enough (people reorganize their entire sites and break old links), but a publicly editable site is just a really bad idea. What are you citing? What it said that day? Because you don't know what it says now. Like ali14 says, just get it from a textbook or something.
    – msouth
    Nov 6, 2017 at 0:58

Depending on the reason for the citation it could be perfectly fine. For example, if you simply want to use the encyclopedia definition to start your introduction:

According to Wikipedia robotics deals with the design, construction, operation, and use of robots,[1] as well as computer systems for their control, sensory feedback, and information processing. The subject of this thesis ...

There is nothing wrong with this (besides maybe the lack of creativity: citing an encyclopedia is used very often in introductions). However, note that Wikipedia cites an original source for this definition under [1], so in this case it is probably preferable to cite the original (the Oxford dictionary in this example).

However, if your purpose is to use wikipedia as a source of factual (scientific) information it is not a good idea.

  • Why would you have information in an academic paper that you didn't consider "factual"?
    – Sneftel
    Oct 18, 2017 at 13:53
  • 3
    Even for introductory statements, I don't think this is appropriate. I would immediately think less of any manuscript that contains the words "according to Wikipedia" because it shows a lack of effort and ignores the common wisdom that one should, in general, not cite Wikipedia. For broad, common knowledge statements like in your example, I would rather see no citation than a citation from Wikipedia. Oct 18, 2017 at 13:54
  • @Sneftel To explain why it is not factual?
    – JAB
    Oct 18, 2017 at 13:55
  • @NuclearWang That says more about you than about the author of the article or wikipedia. You are the one showing "lack of effort" by judging too fast. It is perfectly valid to cite an encyclopedia if you want to refer to the encyclopedic definition of a word. Wikipedia is an encyclopedia - nothing more, nothing less.
    – Louic
    Oct 18, 2017 at 15:57
  • 1
    @louic Literally every answer to this question agrees: do not cite Wikipedia. Instead, cite the source material for the Wikipedia article. There's no reason to skip that step unless you just don't want to bother with the original sources. It would appear I'm judging Wikipedia citations as harshly as anyone else here. Oct 18, 2017 at 17:43

No, you should not. Wikipedia is not a scientific source. The scientific source is the primary peer-reviewed reference that wikipedia's articles cite.

You can (and should) use Wikipedia for a general understanding of a concept, but you should (read and) cite the primary source of the particular concept.

Finally, I would avoid writing "According to X, Y leads to Z" altogether and instead write "Y leads to Z because ...[citeX1, citeX2]". "According to X" is an authoritative argument.


In addition to the other arguments: Citing web sources is a problem, because you do not know if they stay online and if they will be changed in the future. Wikipedia is a bit better than other websites, as they have a clear edit history and citing with date accessed on ... is possible. But you're still relying on the article being still there. For deleted articles there is no (easy to view) history.

  • 1
    It is not "fetched on..", it is "access date" or "date accessed".
    – Coder
    Oct 18, 2017 at 15:50
  • Changed it, thank you. Thought of the practice which is suggested for some references (not citation) to save a copy of the current version of some page to be able to reference to it when the website goes down.
    – allo
    Oct 19, 2017 at 7:40

Do you mean can you cite Wikipedia or can you use the phrase "according to Wikipedia"? Of course citing Wikipedia is not recommended since it can be highly inaccurate [look at e.g. Loudwire's "Wikipedia: Fact or Fiction" YouTube series where bands go through their history on the site and tell the interviewer what is true and what is false].

Using the phrase, "according to Wikipedia..." is of course absolutely fine, though I would suffix your information with a caveat that the citation is unreliable. You could also cite another piece of related information from a more reliable source, and make a reliability comparison.

Ultimately though, it's best to avoid citing Wikipedia altogether. If it is the only place the information you can find is cropping up, then it is most likely false.


hard to imagine someone not knowing the answer to this. 1) go to wikipedia 2) just cite whatever citation wikipedia gives you (hyperlink/bottom of page)

This has been hammered into my head since like 4th grade...


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